[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M72W9GDBR8Q2RMG680FCAB2Y":3,"skill-same:01M72W9GDBR8Q2RMG680FCAB2Y":45},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M72W9GDBR8Q2RMG680FCAB2Y","dna-paint",null,"Prepare and acquire DNA-PAINT samples using the full kinetics, buffer, and bench protocol.","Micro-Claw\u002Fmicroclaw","Micro-Claw","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F320137264?v=4","01M72W9AP162N8788QFVMKTN9H","MicroClaw","01M72W9APY8G657Y1VN8K9S04G",10,"BSD-3-Clause","https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw\u002Fblob\u002Fmain\u002Fmicroclaw\u002Fskills\u002Fdna-paint\u002FSKILL.md",[],[],"2026-11-10 22:39:42",false,"---\nname: dna-paint\ndescription: Prepare and acquire DNA-PAINT samples using the full kinetics, buffer, and bench protocol.\n---\n\n# DNA-PAINT Experiment Protocol — DNA Origami (in vitro), Singleplex\n\nReturned by `load_skill(name=\"dna-paint\")`. `load_skill(name=\"smlm\")` is the\ngeneral SMLM reference and covers DNA-PAINT at planning depth; this document is\nthe depth behind it — kinetics, buffers, strand design, and the full bench\nprocedure. Where the two differ, **this document is the accurate one**.\n\nParameter values below are the published protocol's, measured on the reference\ninstrument in [Schnitzbauer2017]. They are starting points to confirm against\nthe user's own rig, sample and imager stock — never settings to apply unasked.\n\n## Sources\n\nCompiled from two reference papers, cited throughout as:\n\n- **[Schnitzbauer2017]** — Schnitzbauer J, Strauss MT, Schlichthaerle T, Schueder F, Jungmann R.\n  \"Super-resolution microscopy with DNA-PAINT.\" *Nat Protoc.* 2017;12(6):1198–1228.\n  [doi:10.1038\u002Fnprot.2017.024](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fnprot.2017.024) — the detailed, step-by-step\n  DNA-PAINT\u002FPicasso wet-lab protocol; `Step N` below refers to this paper. This is also the\n  primary reference setup this document is modeled on — see [§0](#0-this-experiments-scope).\n- **[Lelek2021]** — Lelek M, Gyparaki MT, Beliu G, Schueder F, Griffié J, Manley S, Jungmann R,\n  Sauer M, Lakadamyali M, Zimmer C. \"Single-molecule localization microscopy.\"\n  *Nat Rev Methods Primers.* 2021;1:39.\n  [doi:10.1038\u002Fs43586-021-00038-x](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs43586-021-00038-x) — broader SMLM primer covering kinetics and general\n  experimental considerations.\n\n---\n\n## 0. This experiment's scope\n\nDecided for this first run:\n\n- **Sample**: DNA origami nanostructures only (in vitro), no cellular\u002Fin situ labeling.\n  This skips immunofixation, antibody-DNA conjugation and cell culture entirely\n  [Schnitzbauer2017, Steps 1–18 apply; Steps 19–33 (in situ) do not].\n- **Multiplexing**: singleplex — one imager species, one acquisition, no Exchange-PAINT fluid\n  exchange and no spectral multiplexing. Simpler chamber, simpler acquisition.\n- **Oxygen scavenging**: the PPT system (PCA\u002FPCD\u002FTrolox) is included from the start — mixed\n  into the imager solution at least 1 h before imaging to improve dye photostability and photon\n  yield [Schnitzbauer2017]. Full ultra-resolution settings (Box 2: 350 ms exposure, 80,000\n  frames, ~4.5 kW\u002Fcm², drift-marker origami) are not adopted here — just the buffer.\n- **Microscope**: any inverted stand with TIRF (or HILO) illumination, a hardware focus\n  lock, a low-noise camera (EMCCD or sCMOS) and an excitation line matching the imager dye.\n  The reference setup in [Schnitzbauer2017, EQUIPMENT] is a Nikon Ti-Eclipse + Perfect Focus\n  System with an Andor iXon Ultra DU-897 EMCCD; its camera\u002Ffocus recommendations transfer to\n  a comparably equipped rig — see [§5](#5-imaging-parameters). **Ask the user what they have\n  rather than assuming this one**, and call `get_system_state` to see what is configured.\n- **Reagents**: all reagents for the experiment are already on hand, so the antibody-labeling\n  and cell-culture reagent lists from the source paper are omitted below.\n\n**Assumption to confirm**: for a 561 nm excitation line, this protocol defaults to the\npaper's 561 nm-compatible reference handle — imager `CTAGATGTAT` labeled with **Cy3B**,\ndocking strand `TTATACATCTA` (the \"P1\" handle) [Schnitzbauer2017, MATERIALS]. If the user's origami\nstructures\u002Fimager stock use a different sequence or dye, swap it in — nothing else in this\nprotocol depends on the specific sequence, only on duplex length and dye\u002Flaser match\n(see [§4](#4-imagerdocking-strand-design)).\n\n**Assumption to confirm**: this protocol includes the full design→fold→purify workflow\n(§6.A–B) for a start from unfolded staples. With folded, purified origami already on hand,\nskip straight to [§6.C](#c-immobilization-on-glass).\n\n---\n\n## 1. Background & purpose\n\nIn DNA-PAINT, transient hybridization of short, dye-labeled **\"imager\" strands** to their\ncomplementary **\"docking\" strands** (here, staple extensions on a DNA origami structure)\ncreates the stochastic fluorescence \"blinking\" needed for single-molecule localization\nmicroscopy (SMLM). This decouples blinking from dye photophysics, so imaging works with\nvirtually any single-molecule-compatible dye and requires no photoswitching buffer\n[Schnitzbauer2017, Introduction; Lelek2021, \"Temporarily binding\" fluorophore class].\n\nWhy DNA origami as the first sample: it's a known-geometry, ground-truth test structure —\ndocking sites sit at defined, designed positions, so imager concentration, exposure time and\nlaser power can be validated against a known answer before ever touching a biological sample\n[Schnitzbauer2017, \"Design and preparation of DNA origami structures\"]. It's also the field's\nstandard calibration\u002Freference structure for exactly this reason\n[Schnitzbauer2017, Fig. 1b].\n\nKey properties motivating DNA-PAINT generally:\n\n- **No photobleaching as a hard limit** — the imager-strand reservoir in solution is\n  effectively infinite, so bleaching of an individual dye has only minimal effect on the\n  overall experiment [Schnitzbauer2017].\n- **Programmable blinking kinetics** — bright time τ_b = 1\u002Fk_off is set by duplex stability\n  (length, GC content, salt\u002Ftemperature); dark time τ_d = 1\u002F(c_i·k_on) is set by imager\n  concentration, with k_on ≈ 10⁶ M⁻¹s⁻¹ typical for DNA-PAINT duplexes [Lelek2021]. A 1-bp\n  increase in duplex length increases τ_b by roughly an order of magnitude [Lelek2021].\n- **~1 nm localization precision \u002F sub-5 nm resolution** has been demonstrated with this\n  approach under optimized (\"ultra-resolution\") conditions [Schnitzbauer2017].\n\nLimitations to keep in mind [Schnitzbauer2017, \"Advantages and limitations\"]: imager strands\nare non-fluorogenic (always present, always slightly fluorescent in solution), so DNA-PAINT\nrequires optical sectioning (TIRF\u002FHILO) to suppress background from unbound imager.\n\n---\n\n## 2. Materials & reagents\n\nCondensed from [Schnitzbauer2017, MATERIALS\u002FREAGENTS\u002FEQUIPMENT], limited to what's needed for\na singleplex in vitro DNA origami experiment (antibody-conjugation and cell-culture reagents\nfrom the source paper are not needed here and are omitted).\n\n### DNA origami folding & purification\n- Staple strands, modified (docking-strand extensions) and unmodified core staples.\n- M13mp18 ssDNA scaffold (p7249).\n- Biotinylated staples (for surface attachment).\n- Tris pH 8.0 (1 M), EDTA pH 8.0 (0.5 M), magnesium (1 M), water — for 10× folding buffer\n  (see [§3](#3-buffer-conditions)).\n- Agarose, 50× TAE buffer, SYBR Safe DNA gel stain, DNA gel loading dye, DNA ladder — for gel\n  purification.\n\n### Immobilization & imaging\n- BSA-biotin (10 mg\u002Fml stock), streptavidin (10 mg\u002Fml stock).\n- Imager strand, dye-labeled (see [§4](#4-imagerdocking-strand-design)).\n- 8 mm-channel custom flow chamber: microscopy slide, high-precision cover glass\n  (No. 1.5H), double-sided adhesive tape, epoxy glue.\n- Isopropanol (cleaning slides\u002Fcoverslips).\n- Tween 20 (buffer component).\n\n### Oxygen-scavenging system (PPT)\n- Protocatechuic acid (PCA), protocatechuate 3,4-dioxygenase (PCD), Trolox, NaOH, methanol —\n  for the 40×\u002F100×\u002F100× stocks mixed 1:1:1 into the imaging buffer (see\n  [§3](#3-buffer-conditions)) [Schnitzbauer2017]. ⚠ PCA and Trolox cause skin\u002Frespiratory\u002Feye\n  irritation — avoid breathing dust\u002Ffumes [Schnitzbauer2017, REAGENTS cautions].\n\n### Equipment\n- Inverted microscope with TIRF (or HILO) illumination and a hardware focus lock.\n- Low-noise camera — EMCCD or back-illuminated sCMOS.\n- Excitation laser matching the imager dye (561 nm for Cy3B). A laser power meter +\n  microscopy-slide thermal power sensor to calibrate actual power **density at the sample\n  plane**, since rated laser output is not the relevant number (see\n  [§5](#5-imaging-parameters)) [Schnitzbauer2017, EQUIPMENT SETUP].\n- Micro-Manager acquisition software [Schnitzbauer2017] — driven through Microclaw\u002Fpycro-manager\n  here.\n- Picasso (jungmannlab.org \u002F github.com\u002Fjungmannlab\u002Fpicasso) for origami design, localization,\n  rendering and drift correction [Schnitzbauer2017].\n- Thermocycler (for annealing), gel chamber + power supply + blue-light transilluminator +\n  gel imager (for purification\u002FQC).\n\n---\n\n## 3. Buffer conditions\n\nRecipes from [Schnitzbauer2017, REAGENT SETUP], limited to the in vitro subset:\n\n| Buffer | Composition | Storage |\n|---|---|---|\n| Buffer A | 10 mM Tris-HCl, 100 mM NaCl, pH 8.0 | RT, 6 months |\n| Buffer A+ | Buffer A + 0.05% (vol\u002Fvol) Tween 20 | RT, 6 months |\n| Buffer B | 5 mM Tris-HCl, 10 mM MgCl₂, 1 mM EDTA, pH 8.0 | RT, 6 months |\n| Buffer B+ | Buffer B + 0.05% (vol\u002Fvol) Tween 20 | RT, 6 months |\n| 10× folding buffer | 125 mM MgCl₂, 100 mM Tris, 10 mM EDTA, pH 8.0 | RT, 6 months |\n| Gel buffer | 1× TAE | RT, 1 year |\n| Gel running buffer | 1× TAE + 12.5 mM MgCl₂ | RT, 1 year |\n| BSA-biotin stock | 10 mg\u002Fml BSA-biotin in Buffer A; 20 µl aliquots | −20 °C, 6 months |\n| BSA-biotin solution | 1 mg\u002Fml BSA-biotin in Buffer A+ | 4 °C, 3 days (fresh) |\n| Streptavidin stock | 10 mg\u002Fml streptavidin in Buffer A; 10 µl aliquots | −20 °C, 6 months |\n| Streptavidin solution | 0.5 mg\u002Fml streptavidin in Buffer A+ | 4 °C, 3 days (fresh) |\n| 100× Trolox | 100 mg Trolox + 430 µl methanol + 345 µl 1 M NaOH in 3.2 ml H₂O; 20 µl aliquots | −20 °C, 6 months |\n| 40× PCA | 154 mg PCA in 10 ml water, pH 9.0 (NaOH); 20 µl aliquots | −20 °C, 6 months |\n| 100× PCD | 9.3 mg PCD in 13.3 ml (50% glycerol, 50 mM KCl, 1 mM EDTA, 100 mM Tris-HCl pH 8.0); 20 µl aliquots | −20 °C, 6 months |\n| **PPT oxygen-scavenging mix** | 1:1:1 ratio of 1× PCA : 1× PCD : 1× Trolox, mixed into imager solution **≥1 h before imaging** | fresh |\n| **Imager solution (in vitro, with PPT)** | 1× Buffer B+ + 1× PCA + 1× PCD + 1× Trolox + fluorophore-labeled DNA strand, 100 pM–10 nM (start ~5 nM for a 12-site structure) | fresh |\n\nPPT improves fluorophore photostability and therefore photon yield per binding event\n[Schnitzbauer2017, Box 2] — worth having in from the start rather than only for a dedicated\nultra-resolution attempt, since it costs little beyond a 1 h pre-mix wait.\n\nImager concentration is the most consequential tuning knob: too low under-samples the\nstructure; too high raises unbound-imager background and causes \"cross-talk\" localizations\nwhere two nearby sites bind simultaneously and get fitted as a single false spot in between\n[Schnitzbauer2017, Fig. 7d]. Picasso's `Simulate` module can screen concentration\u002Fexposure\ncombinations in silico before committing microscope time [Schnitzbauer2017, Box 3].\n\n---\n\n## 4. Imager\u002Fdocking strand design\n\n- Docking and imager strands are short complementary single-stranded DNA oligos, typically\n  **8–10 nucleotides** [Schnitzbauer2017]. Here, the docking strand is a staple extension on\n  the origami; the imager strand carries the dye and diffuses freely until it transiently\n  binds.\n- **Default handle for this protocol** (561 nm \u002F Cy3B match — confirm against the user's\n  actual stock, see [§0](#0-this-experiments-scope)):\n  - Docking (staple extension): `TTATACATCTA`\n  - Imager: `CTAGATGTAT` labeled with **Cy3B**, excited at 561 nm\n  [Schnitzbauer2017, MATERIALS; this is the same P1 sequence used for the calibration\n  experiments on the paper's own TIRF setup].\n- **Duplex length sets the bright\u002FON time** (τ_b = 1\u002Fk_off): a typical 9-bp duplex (like P1\n  above) gives τ_b ≈ 500 ms. Each added base pair increases τ_b roughly 10×; each removed base\n  pair decreases it roughly 10× [Lelek2021]. Camera exposure time should be matched to τ_b\n  [Schnitzbauer2017, \"Data acquisition\"].\n- **Imager concentration sets the dark\u002FOFF time** (τ_d = 1\u002F(c_i·k_on)), k_on ≈ 10⁶ M⁻¹s⁻¹\n  typical [Lelek2021].\n- **Estimating required imaging time**: at 10 nM imager and k_on ≈ 10⁶ M⁻¹s⁻¹,\n  τ_d = (k_on·c)⁻¹ ≈ 100 s. For a ~98% probability that any given site is visited at least once\n  (P = 1 − e^(−t\u002Fτ_d)), total imaging time t ≈ 4·τ_d ≈ 400 s; [Schnitzbauer2017] recommends\n  **~33 min total** to get multiple binding events per site (needed for decent image quality,\n  not just \"visited once\").\n- Docking-site layout (hexagonal lattice, 5 nm spacing) is designed in `Picasso: Design`, which\n  also auto-generates staple order lists and pipetting schemes [Schnitzbauer2017, Steps 1–5].\n\n---\n\n## 5. Imaging parameters\n\nThe published values, measured on the paper's reference instrument\n[Schnitzbauer2017, EQUIPMENT]. They transfer to a comparably equipped rig — TIRF, hardware\nfocus lock, low-noise camera — and each is a starting point to confirm, not a setting to\napply unasked.\n\n| Parameter | Recommendation | Source |\n|---|---|---|\n| Exposure time | ~300 ms (paper's in vitro Buffer B+ example, matched to the P1 9-bp duplex τ_b ≈ 500 ms) | [Schnitzbauer2017, Step 37] |\n| Frame interval | 0 ms (back-to-back frames) | [Schnitzbauer2017, Step 47] |\n| Frame count | Start at 7,500; typical full datasets 10,000–100,000+ depending on desired sampling | [Schnitzbauer2017, Step 46] |\n| Pre-focus laser power density | ~0.25 kW\u002Fcm² at the sample plane | [Schnitzbauer2017, Step 41] |\n| Imaging laser power density | ~2.5–6 kW\u002Fcm² at the sample plane (561 nm\u002FCy3B reference range: 1–6 kW\u002Fcm²); raise until bright time (τ_b) starts to *decrease* — that's imager bleaching while bound, and is the practical ceiling | [Schnitzbauer2017, Step 43; Table 1 troubleshooting] |\n| **Power density calibration** | **Power density at the sample plane is what matters, not rated laser output** — measure it directly with a microscopy-slide thermal power sensor before picking a % laser power setting. A watt-class source runs at a small fraction of full power for this, so a % setting carried over from another experiment is not evidence of anything. | [Schnitzbauer2017, EQUIPMENT SETUP, \"Power density calibration\"] |\n| Illumination mode | TIRF where available — best background suppression for surface-immobilized origami; HILO otherwise | [Schnitzbauer2017, Step 44] |\n| Focus | Engage the hardware focus lock for prefocus and for the run (`set_focus_lock`) — the paper explicitly recommends a hardware lock over software refocusing | [Schnitzbauer2017, Step 42, CRITICAL STEP] |\n| Camera settings | EMCCD reference values: Output amplifier: Conventional; ROI: Full Image; Frame Transfer: On; PixelType: 16-bit; ReadMode: Image; shutters: Open; readout mode = lowest-noise frequency whose readout time doesn't exceed the exposure time. On an sCMOS, the equivalent choice is the lowest-noise readout mode that sustains the frame rate | [Schnitzbauer2017, Steps 39, 45] |\n| Photon conversion (for `Picasso: Localize`) | EM Gain, Baseline, Sensitivity, Quantum Efficiency set per the camera's actual specs — set EM Gain = 1 when using conventional (non-EM) amplification | [Schnitzbauer2017, Step 54] |\n| Equilibration before acquiring | 5–15 min on the microscope before starting, to let thermal\u002Fmechanical drift settle | [Schnitzbauer2017, Table 1 troubleshooting] |\n\n---\n\n## 6. Step-by-step procedure\n\n### A. Design & fold DNA origami *(skip if structures are already folded)*\n\n[Schnitzbauer2017, Steps 1–9]:\n\n1. In `Picasso: Design`, lay out docking sites on the hexagonal lattice canvas (5 nm\n   spacing) and assign the extension sequence (§4) (Steps 1–4).\n2. Export the staple order list (`Get plates`) and order staples, plus the biotinylated\n   surface-attachment staples (Step 5). *(Pause point: 2–10 working days for synthesis.)*\n3. Pool staples per the auto-generated pipetting scheme into stock mixes (Steps 6–7).\n4. Fold via thermal annealing: 80 °C hold, then ramp 60 °C → 4 °C at ~3 min 12 s\u002F°C over\n   ~57 cycles, hold at 4 °C (Step 9). *(Pause point: store folded structures at 4 °C up to\n   1 week, or −20 °C long-term.)*\n\n### B. Purify\n\n[Schnitzbauer2017, Steps 10–17]:\n\n1. Run a 1.5% agarose gel (gel buffer + MgCl₂ + SYBR Safe), 90 V, 90 min, at 4 °C\u002Fon ice\n   (Steps 11–15).\n2. Image the gel; excise the origami band (a single sharp band, distinct from and slightly\n   shifted relative to bare scaffold) (Step 16–17).\n3. Recover via Freeze 'N Squeeze spin column, 6 min at 1,000g, 4 °C (Step 17).\n   *(Pause point: store purified origami at 4 °C for 1 week, or −20 °C for long-term.)*\n\n### C. Immobilization on glass\n\n[Schnitzbauer2017, Step 18, option A — sealed custom chamber]:\n\n1. Clean a microscopy slide and coverslip with isopropanol.\n2. Build an ~8 mm-wide flow chamber with two strips of double-sided tape between slide and\n   coverslip (~20–30 µl channel volume).\n3. BSA-biotin solution (1 mg\u002Fml), 2 min → wash with Buffer A+.\n4. Streptavidin solution (0.5 mg\u002Fml), 2 min → wash with Buffer A+, then Buffer B+.\n5. Origami solution (~5 nM for a 12-site structure, diluted in Buffer B+), 2 min → wash with\n   Buffer B+.\n6. Add imager solution with PPT (§3\u002F§4). ⚠ Mix the PCA\u002FPCD\u002FTrolox into the imager solution\n   **at least 1 h before this step** — prepare it while folding\u002Fpurifying, not last-minute\n   [Schnitzbauer2017].\n7. Seal the chamber with epoxy glue; wait ~15 min for it to dry before mounting.\n   ⚠ Wait until fully dry — uncured epoxy can contaminate the objective\n   [Schnitzbauer2017, Step 18A, CRITICAL STEP].\n\n### D. Data acquisition\n\n[Schnitzbauer2017, Steps 34–48], with the microclaw call for each step where there is one:\n\n1. Mount the sample; raise the objective until immersion oil contacts the coverslip\n   (Step 34).\n2. With Micro-Manager running, confirm the configured camera and channel with\n   `get_system_state` (Step 36).\n3. Set exposure ≈ 300 ms and the camera parameters from [§5](#5-imaging-parameters)\n   (Steps 37, 39). Exposure travels with the acquisition call — `run_timelapse(exposure_ms=300)`\n   — rather than being set separately.\n4. Go live, autostretch contrast — confirm you see background noise only (Step 40).\n5. Open the excitation shutter at low power (~0.25 kW\u002Fcm² at the sample — calibrate against\n   the power meter, not the laser dial) and focus, then engage the hardware focus lock\n   (`run_autofocus`, then `set_focus_lock(enabled=true)`) (Steps 41–42).\n6. Raise power to imaging density (~2.5–6 kW\u002Fcm²); you should now see individual blinking,\n   diffraction-limited spots (Step 43).\n7. Adjust the TIRF incident angle for best SNR (Step 44).\n8. Set EMCCD readout mode for lowest noise consistent with the exposure time (Step 45).\n9. Acquire with `run_timelapse(n_frames=7500, interval_s=0, exposure_ms=300)` — `interval_s=0`\n   is back-to-back frames, and the dataset is saved to disk by the acquisition itself\n   (Steps 46–47). Per-frame hardware control instead requires distinct consecutive\n   `int(k * interval_s * 1000.0)` deadlines throughout the frame count. Equal truncated\n   millisecond deadlines allow hardware-sequencing with no software between exposures;\n   Microclaw refuses such fixed plans before acquisition. Observation-only bursts remain allowed.\n10. Let the sample equilibrate 5–15 min, then acquire (Step 48).\n\n### E. Reconstruction (pointer)\n\nFull detail in [Schnitzbauer2017, Steps 50–60]; summary:\n\n0. Export the raw stack for the localization software with `export_dataset_as_tiff`.\n   Microclaw does not fit localizations; everything below runs outside it.\n1. `Picasso: Localize` — identify and fit single-molecule spots (box side ≈ 6σ_PSF + 1; set a\n   minimum net-gradient threshold; set the camera's photon-conversion parameters from\n   [§5](#5-imaging-parameters)) (Steps 50–55).\n2. `Picasso: Render` — render the super-resolution image (Steps 56–58).\n3. Drift correction — redundant cross-correlation (RCC) first; for higher resolution, follow\n   with fiducial\u002Forigami-based correction (Steps 59–60).\n4. Since the docking-site layout is known by design, compare the reconstructed image against\n   the designed pattern as a sanity check [Schnitzbauer2017, \"Anticipated results\"].\n","# DNA-PAINT Experiment Protocol — DNA Origami (in vitro), Singleplex\n\nReturned by `load_skill(name=\"dna-paint\")`. `load_skill(name=\"smlm\")` is the\ngeneral SMLM reference and covers DNA-PAINT at planning depth; this document is\nthe depth behind it — kinetics, buffers, strand design, and the full bench\nprocedure. Where the two differ, **this document is the accurate one**.\n\nParameter values below are the published protocol's, measured on the reference\ninstrument in [Schnitzbauer2017]. They are starting points to confirm against\nthe user's own rig, sample and imager stock — never settings to apply unasked.\n\n## Sources\n\nCompiled from two reference papers, cited throughout as:\n\n- **[Schnitzbauer2017]** — Schnitzbauer J, Strauss MT, Schlichthaerle T, Schueder F, Jungmann R.\n  \"Super-resolution microscopy with DNA-PAINT.\" *Nat Protoc.* 2017;12(6):1198–1228.\n  [doi:10.1038\u002Fnprot.2017.024](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fnprot.2017.024) — the detailed, step-by-step\n  DNA-PAINT\u002FPicasso wet-lab protocol; `Step N` below refers to this paper. This is also the\n  primary reference setup this document is modeled on — see [§0](#0-this-experiments-scope).\n- **[Lelek2021]** — Lelek M, Gyparaki MT, Beliu G, Schueder F, Griffié J, Manley S, Jungmann R,\n  Sauer M, Lakadamyali M, Zimmer C. \"Single-molecule localization microscopy.\"\n  *Nat Rev Methods Primers.* 2021;1:39.\n  [doi:10.1038\u002Fs43586-021-00038-x](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs43586-021-00038-x) — broader SMLM primer covering kinetics and general\n  experimental considerations.\n\n---\n\n## 0. This experiment's scope\n\nDecided for this first run:\n\n- **Sample**: DNA origami nanostructures only (in vitro), no cellular\u002Fin situ labeling.\n  This skips immunofixation, antibody-DNA conjugation and cell culture entirely\n  [Schnitzbauer2017, Steps 1–18 apply; Steps 19–33 (in situ) do not].\n- **Multiplexing**: singleplex — one imager species, one acquisition, no Exchange-PAINT fluid\n  exchange and no spectral multiplexing. Simpler chamber, simpler acquisition.\n- **Oxygen scavenging**: the PPT system (PCA\u002FPCD\u002FTrolox) is included from the start — mixed\n  into the imager solution at least 1 h before imaging to improve dye photostability and photon\n  yield [Schnitzbauer2017]. Full ultra-resolution settings (Box 2: 350 ms exposure, 80,000\n  frames, ~4.5 kW\u002Fcm², drift-marker origami) are not adopted here — just the buffer.\n- **Microscope**: any inverted stand with TIRF (or HILO) illumination, a hardware focus\n  lock, a low-noise camera (EMCCD or sCMOS) and an excitation line matching the imager dye.\n  The reference setup in [Schnitzbauer2017, EQUIPMENT] is a Nikon Ti-Eclipse + Perfect Focus\n  System with an Andor iXon Ultra DU-897 EMCCD; its camera\u002Ffocus recommendations transfer to\n  a comparably equipped rig — see [§5](#5-imaging-parameters). **Ask the user what they have\n  rather than assuming this one**, and call `get_system_state` to see what is configured.\n- **Reagents**: all reagents for the experiment are already on hand, so the antibody-labeling\n  and cell-culture reagent lists from the source paper are omitted below.\n\n**Assumption to confirm**: for a 561 nm excitation line, this protocol defaults to the\npaper's 561 nm-compatible reference handle — imager `CTAGATGTAT` labeled with **Cy3B**,\ndocking strand `TTATACATCTA` (the \"P1\" handle) [Schnitzbauer2017, MATERIALS]. If the user's origami\nstructures\u002Fimager stock use a different sequence or dye, swap it in — nothing else in this\nprotocol depends on the specific sequence, only on duplex length and dye\u002Flaser match\n(see [§4](#4-imagerdocking-strand-design)).\n\n**Assumption to confirm**: this protocol includes the full design→fold→purify workflow\n(§6.A–B) for a start from unfolded staples. With folded, purified origami already on hand,\nskip straight to [§6.C](#c-immobilization-on-glass).\n\n---\n\n## 1. Background & purpose\n\nIn DNA-PAINT, transient hybridization of short, dye-labeled **\"imager\" strands** to their\ncomplementary **\"docking\" strands** (here, staple extensions on a DNA origami structure)\ncreates the stochastic fluorescence \"blinking\" needed for single-molecule localization\nmicroscopy (SMLM). This decouples blinking from dye photophysics, so imaging works with\nvirtually any single-molecule-compatible dye and requires no photoswitching buffer\n[Schnitzbauer2017, Introduction; Lelek2021, \"Temporarily binding\" fluorophore class].\n\nWhy DNA origami as the first sample: it's a known-geometry, ground-truth test structure —\ndocking sites sit at defined, designed positions, so imager concentration, exposure time and\nlaser power can be validated against a known answer before ever touching a biological sample\n[Schnitzbauer2017, \"Design and preparation of DNA origami structures\"]. It's also the field's\nstandard calibration\u002Freference structure for exactly this reason\n[Schnitzbauer2017, Fig. 1b].\n\nKey properties motivating DNA-PAINT generally:\n\n- **No photobleaching as a hard limit** — the imager-strand reservoir in solution is\n  effectively infinite, so bleaching of an individual dye has only minimal effect on the\n  overall experiment [Schnitzbauer2017].\n- **Programmable blinking kinetics** — bright time τ_b = 1\u002Fk_off is set by duplex stability\n  (length, GC content, salt\u002Ftemperature); dark time τ_d = 1\u002F(c_i·k_on) is set by imager\n  concentration, with k_on ≈ 10⁶ M⁻¹s⁻¹ typical for DNA-PAINT duplexes [Lelek2021]. A 1-bp\n  increase in duplex length increases τ_b by roughly an order of magnitude [Lelek2021].\n- **~1 nm localization precision \u002F sub-5 nm resolution** has been demonstrated with this\n  approach under optimized (\"ultra-resolution\") conditions [Schnitzbauer2017].\n\nLimitations to keep in mind [Schnitzbauer2017, \"Advantages and limitations\"]: imager strands\nare non-fluorogenic (always present, always slightly fluorescent in solution), so DNA-PAINT\nrequires optical sectioning (TIRF\u002FHILO) to suppress background from unbound imager.\n\n---\n\n## 2. Materials & reagents\n\nCondensed from [Schnitzbauer2017, MATERIALS\u002FREAGENTS\u002FEQUIPMENT], limited to what's needed for\na singleplex in vitro DNA origami experiment (antibody-conjugation and cell-culture reagents\nfrom the source paper are not needed here and are omitted).\n\n### DNA origami folding & purification\n- Staple strands, modified (docking-strand extensions) and unmodified core staples.\n- M13mp18 ssDNA scaffold (p7249).\n- Biotinylated staples (for surface attachment).\n- Tris pH 8.0 (1 M), EDTA pH 8.0 (0.5 M), magnesium (1 M), water — for 10× folding buffer\n  (see [§3](#3-buffer-conditions)).\n- Agarose, 50× TAE buffer, SYBR Safe DNA gel stain, DNA gel loading dye, DNA ladder — for gel\n  purification.\n\n### Immobilization & imaging\n- BSA-biotin (10 mg\u002Fml stock), streptavidin (10 mg\u002Fml stock).\n- Imager strand, dye-labeled (see [§4](#4-imagerdocking-strand-design)).\n- 8 mm-channel custom flow chamber: microscopy slide, high-precision cover glass\n  (No. 1.5H), double-sided adhesive tape, epoxy glue.\n- Isopropanol (cleaning slides\u002Fcoverslips).\n- Tween 20 (buffer component).\n\n### Oxygen-scavenging system (PPT)\n- Protocatechuic acid (PCA), protocatechuate 3,4-dioxygenase (PCD), Trolox, NaOH, methanol —\n  for the 40×\u002F100×\u002F100× stocks mixed 1:1:1 into the imaging buffer (see\n  [§3](#3-buffer-conditions)) [Schnitzbauer2017]. ⚠ PCA and Trolox cause skin\u002Frespiratory\u002Feye\n  irritation — avoid breathing dust\u002Ffumes [Schnitzbauer2017, REAGENTS cautions].\n\n### Equipment\n- Inverted microscope with TIRF (or HILO) illumination and a hardware focus lock.\n- Low-noise camera — EMCCD or back-illuminated sCMOS.\n- Excitation laser matching the imager dye (561 nm for Cy3B). A laser power meter +\n  microscopy-slide thermal power sensor to calibrate actual power **density at the sample\n  plane**, since rated laser output is not the relevant number (see\n  [§5](#5-imaging-parameters)) [Schnitzbauer2017, EQUIPMENT SETUP].\n- Micro-Manager acquisition software [Schnitzbauer2017] — driven through Microclaw\u002Fpycro-manager\n  here.\n- Picasso (jungmannlab.org \u002F github.com\u002Fjungmannlab\u002Fpicasso) for origami design, localization,\n  rendering and drift correction [Schnitzbauer2017].\n- Thermocycler (for annealing), gel chamber + power supply + blue-light transilluminator +\n  gel imager (for purification\u002FQC).\n\n---\n\n## 3. Buffer conditions\n\nRecipes from [Schnitzbauer2017, REAGENT SETUP], limited to the in vitro subset:\n\n| Buffer | Composition | Storage |\n|---|---|---|\n| Buffer A | 10 mM Tris-HCl, 100 mM NaCl, pH 8.0 | RT, 6 months |\n| Buffer A+ | Buffer A + 0.05% (vol\u002Fvol) Tween 20 | RT, 6 months |\n| Buffer B | 5 mM Tris-HCl, 10 mM MgCl₂, 1 mM EDTA, pH 8.0 | RT, 6 months |\n| Buffer B+ | Buffer B + 0.05% (vol\u002Fvol) Tween 20 | RT, 6 months |\n| 10× folding buffer | 125 mM MgCl₂, 100 mM Tris, 10 mM EDTA, pH 8.0 | RT, 6 months |\n| Gel buffer | 1× TAE | RT, 1 year |\n| Gel running buffer | 1× TAE + 12.5 mM MgCl₂ | RT, 1 year |\n| BSA-biotin stock | 10 mg\u002Fml BSA-biotin in Buffer A; 20 µl aliquots | −20 °C, 6 months |\n| BSA-biotin solution | 1 mg\u002Fml BSA-biotin in Buffer A+ | 4 °C, 3 days (fresh) |\n| Streptavidin stock | 10 mg\u002Fml streptavidin in Buffer A; 10 µl aliquots | −20 °C, 6 months |\n| Streptavidin solution | 0.5 mg\u002Fml streptavidin in Buffer A+ | 4 °C, 3 days (fresh) |\n| 100× Trolox | 100 mg Trolox + 430 µl methanol + 345 µl 1 M NaOH in 3.2 ml H₂O; 20 µl aliquots | −20 °C, 6 months |\n| 40× PCA | 154 mg PCA in 10 ml water, pH 9.0 (NaOH); 20 µl aliquots | −20 °C, 6 months |\n| 100× PCD | 9.3 mg PCD in 13.3 ml (50% glycerol, 50 mM KCl, 1 mM EDTA, 100 mM Tris-HCl pH 8.0); 20 µl aliquots | −20 °C, 6 months |\n| **PPT oxygen-scavenging mix** | 1:1:1 ratio of 1× PCA : 1× PCD : 1× Trolox, mixed into imager solution **≥1 h before imaging** | fresh |\n| **Imager solution (in vitro, with PPT)** | 1× Buffer B+ + 1× PCA + 1× PCD + 1× Trolox + fluorophore-labeled DNA strand, 100 pM–10 nM (start ~5 nM for a 12-site structure) | fresh |\n\nPPT improves fluorophore photostability and therefore photon yield per binding event\n[Schnitzbauer2017, Box 2] — worth having in from the start rather than only for a dedicated\nultra-resolution attempt, since it costs little beyond a 1 h pre-mix wait.\n\nImager concentration is the most consequential tuning knob: too low under-samples the\nstructure; too high raises unbound-imager background and causes \"cross-talk\" localizations\nwhere two nearby sites bind simultaneously and get fitted as a single false spot in between\n[Schnitzbauer2017, Fig. 7d]. Picasso's `Simulate` module can screen concentration\u002Fexposure\ncombinations in silico before committing microscope time [Schnitzbauer2017, Box 3].\n\n---\n\n## 4. Imager\u002Fdocking strand design\n\n- Docking and imager strands are short complementary single-stranded DNA oligos, typically\n  **8–10 nucleotides** [Schnitzbauer2017]. Here, the docking strand is a staple extension on\n  the origami; the imager strand carries the dye and diffuses freely until it transiently\n  binds.\n- **Default handle for this protocol** (561 nm \u002F Cy3B match — confirm against the user's\n  actual stock, see [§0](#0-this-experiments-scope)):\n  - Docking (staple extension): `TTATACATCTA`\n  - Imager: `CTAGATGTAT` labeled with **Cy3B**, excited at 561 nm\n  [Schnitzbauer2017, MATERIALS; this is the same P1 sequence used for the calibration\n  experiments on the paper's own TIRF setup].\n- **Duplex length sets the bright\u002FON time** (τ_b = 1\u002Fk_off): a typical 9-bp duplex (like P1\n  above) gives τ_b ≈ 500 ms. Each added base pair increases τ_b roughly 10×; each removed base\n  pair decreases it roughly 10× [Lelek2021]. Camera exposure time should be matched to τ_b\n  [Schnitzbauer2017, \"Data acquisition\"].\n- **Imager concentration sets the dark\u002FOFF time** (τ_d = 1\u002F(c_i·k_on)), k_on ≈ 10⁶ M⁻¹s⁻¹\n  typical [Lelek2021].\n- **Estimating required imaging time**: at 10 nM imager and k_on ≈ 10⁶ M⁻¹s⁻¹,\n  τ_d = (k_on·c)⁻¹ ≈ 100 s. For a ~98% probability that any given site is visited at least once\n  (P = 1 − e^(−t\u002Fτ_d)), total imaging time t ≈ 4·τ_d ≈ 400 s; [Schnitzbauer2017] recommends\n  **~33 min total** to get multiple binding events per site (needed for decent image quality,\n  not just \"visited once\").\n- Docking-site layout (hexagonal lattice, 5 nm spacing) is designed in `Picasso: Design`, which\n  also auto-generates staple order lists and pipetting schemes [Schnitzbauer2017, Steps 1–5].\n\n---\n\n## 5. Imaging parameters\n\nThe published values, measured on the paper's reference instrument\n[Schnitzbauer2017, EQUIPMENT]. They transfer to a comparably equipped rig — TIRF, hardware\nfocus lock, low-noise camera — and each is a starting point to confirm, not a setting to\napply unasked.\n\n| Parameter | Recommendation | Source |\n|---|---|---|\n| Exposure time | ~300 ms (paper's in vitro Buffer B+ example, matched to the P1 9-bp duplex τ_b ≈ 500 ms) | [Schnitzbauer2017, Step 37] |\n| Frame interval | 0 ms (back-to-back frames) | [Schnitzbauer2017, Step 47] |\n| Frame count | Start at 7,500; typical full datasets 10,000–100,000+ depending on desired sampling | [Schnitzbauer2017, Step 46] |\n| Pre-focus laser power density | ~0.25 kW\u002Fcm² at the sample plane | [Schnitzbauer2017, Step 41] |\n| Imaging laser power density | ~2.5–6 kW\u002Fcm² at the sample plane (561 nm\u002FCy3B reference range: 1–6 kW\u002Fcm²); raise until bright time (τ_b) starts to *decrease* — that's imager bleaching while bound, and is the practical ceiling | [Schnitzbauer2017, Step 43; Table 1 troubleshooting] |\n| **Power density calibration** | **Power density at the sample plane is what matters, not rated laser output** — measure it directly with a microscopy-slide thermal power sensor before picking a % laser power setting. A watt-class source runs at a small fraction of full power for this, so a % setting carried over from another experiment is not evidence of anything. | [Schnitzbauer2017, EQUIPMENT SETUP, \"Power density calibration\"] |\n| Illumination mode | TIRF where available — best background suppression for surface-immobilized origami; HILO otherwise | [Schnitzbauer2017, Step 44] |\n| Focus | Engage the hardware focus lock for prefocus and for the run (`set_focus_lock`) — the paper explicitly recommends a hardware lock over software refocusing | [Schnitzbauer2017, Step 42, CRITICAL STEP] |\n| Camera settings | EMCCD reference values: Output amplifier: Conventional; ROI: Full Image; Frame Transfer: On; PixelType: 16-bit; ReadMode: Image; shutters: Open; readout mode = lowest-noise frequency whose readout time doesn't exceed the exposure time. On an sCMOS, the equivalent choice is the lowest-noise readout mode that sustains the frame rate | [Schnitzbauer2017, Steps 39, 45] |\n| Photon conversion (for `Picasso: Localize`) | EM Gain, Baseline, Sensitivity, Quantum Efficiency set per the camera's actual specs — set EM Gain = 1 when using conventional (non-EM) amplification | [Schnitzbauer2017, Step 54] |\n| Equilibration before acquiring | 5–15 min on the microscope before starting, to let thermal\u002Fmechanical drift settle | [Schnitzbauer2017, Table 1 troubleshooting] |\n\n---\n\n## 6. Step-by-step procedure\n\n### A. Design & fold DNA origami *(skip if structures are already folded)*\n\n[Schnitzbauer2017, Steps 1–9]:\n\n1. In `Picasso: Design`, lay out docking sites on the hexagonal lattice canvas (5 nm\n   spacing) and assign the extension sequence (§4) (Steps 1–4).\n2. Export the staple order list (`Get plates`) and order staples, plus the biotinylated\n   surface-attachment staples (Step 5). *(Pause point: 2–10 working days for synthesis.)*\n3. Pool staples per the auto-generated pipetting scheme into stock mixes (Steps 6–7).\n4. Fold via thermal annealing: 80 °C hold, then ramp 60 °C → 4 °C at ~3 min 12 s\u002F°C over\n   ~57 cycles, hold at 4 °C (Step 9). *(Pause point: store folded structures at 4 °C up to\n   1 week, or −20 °C long-term.)*\n\n### B. Purify\n\n[Schnitzbauer2017, Steps 10–17]:\n\n1. Run a 1.5% agarose gel (gel buffer + MgCl₂ + SYBR Safe), 90 V, 90 min, at 4 °C\u002Fon ice\n   (Steps 11–15).\n2. Image the gel; excise the origami band (a single sharp band, distinct from and slightly\n   shifted relative to bare scaffold) (Step 16–17).\n3. Recover via Freeze 'N Squeeze spin column, 6 min at 1,000g, 4 °C (Step 17).\n   *(Pause point: store purified origami at 4 °C for 1 week, or −20 °C for long-term.)*\n\n### C. Immobilization on glass\n\n[Schnitzbauer2017, Step 18, option A — sealed custom chamber]:\n\n1. Clean a microscopy slide and coverslip with isopropanol.\n2. Build an ~8 mm-wide flow chamber with two strips of double-sided tape between slide and\n   coverslip (~20–30 µl channel volume).\n3. BSA-biotin solution (1 mg\u002Fml), 2 min → wash with Buffer A+.\n4. Streptavidin solution (0.5 mg\u002Fml), 2 min → wash with Buffer A+, then Buffer B+.\n5. Origami solution (~5 nM for a 12-site structure, diluted in Buffer B+), 2 min → wash with\n   Buffer B+.\n6. Add imager solution with PPT (§3\u002F§4). ⚠ Mix the PCA\u002FPCD\u002FTrolox into the imager solution\n   **at least 1 h before this step** — prepare it while folding\u002Fpurifying, not last-minute\n   [Schnitzbauer2017].\n7. Seal the chamber with epoxy glue; wait ~15 min for it to dry before mounting.\n   ⚠ Wait until fully dry — uncured epoxy can contaminate the objective\n   [Schnitzbauer2017, Step 18A, CRITICAL STEP].\n\n### D. Data acquisition\n\n[Schnitzbauer2017, Steps 34–48], with the microclaw call for each step where there is one:\n\n1. Mount the sample; raise the objective until immersion oil contacts the coverslip\n   (Step 34).\n2. With Micro-Manager running, confirm the configured camera and channel with\n   `get_system_state` (Step 36).\n3. Set exposure ≈ 300 ms and the camera parameters from [§5](#5-imaging-parameters)\n   (Steps 37, 39). Exposure travels with the acquisition call — `run_timelapse(exposure_ms=300)`\n   — rather than being set separately.\n4. Go live, autostretch contrast — confirm you see background noise only (Step 40).\n5. Open the excitation shutter at low power (~0.25 kW\u002Fcm² at the sample — calibrate against\n   the power meter, not the laser dial) and focus, then engage the hardware focus lock\n   (`run_autofocus`, then `set_focus_lock(enabled=true)`) (Steps 41–42).\n6. Raise power to imaging density (~2.5–6 kW\u002Fcm²); you should now see individual blinking,\n   diffraction-limited spots (Step 43).\n7. Adjust the TIRF incident angle for best SNR (Step 44).\n8. Set EMCCD readout mode for lowest noise consistent with the exposure time (Step 45).\n9. Acquire with `run_timelapse(n_frames=7500, interval_s=0, exposure_ms=300)` — `interval_s=0`\n   is back-to-back frames, and the dataset is saved to disk by the acquisition itself\n   (Steps 46–47). Per-frame hardware control instead requires distinct consecutive\n   `int(k * interval_s * 1000.0)` deadlines throughout the frame count. Equal truncated\n   millisecond deadlines allow hardware-sequencing with no software between exposures;\n   Microclaw refuses such fixed plans before acquisition. Observation-only bursts remain allowed.\n10. Let the sample equilibrate 5–15 min, then acquire (Step 48).\n\n### E. Reconstruction (pointer)\n\nFull detail in [Schnitzbauer2017, Steps 50–60]; summary:\n\n0. Export the raw stack for the localization software with `export_dataset_as_tiff`.\n   Microclaw does not fit localizations; everything below runs outside it.\n1. `Picasso: Localize` — identify and fit single-molecule spots (box side ≈ 6σ_PSF + 1; set a\n   minimum net-gradient threshold; set the camera's photon-conversion parameters from\n   [§5](#5-imaging-parameters)) (Steps 50–55).\n2. `Picasso: Render` — render the super-resolution image (Steps 56–58).\n3. Drift correction — redundant cross-correlation (RCC) first; for higher resolution, follow\n   with fiducial\u002Forigami-based correction (Steps 59–60).\n4. Since the docking-site layout is known by design, compare the reconstructed image against\n   the designed pattern as a sanity check [Schnitzbauer2017, \"Anticipated results\"].\n",{"name":5,"description":7},"main","bb3e123c0e5ce6b52cb0e1a27d5e52bce5f3b045","microclaw\u002Fskills\u002Fdna-paint",1,"https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw","https:\u002F\u002Fgithub.com\u002FMicro-Claw","2026-10-20 00:35:52",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add Micro-Claw\u002Fmicroclaw --skill dna-paint",{"tool":37,"command":38},"bunx","bunx skills add Micro-Claw\u002Fmicroclaw --skill dna-paint",{"tool":40,"command":41},"pnpm","pnpm dlx skills add Micro-Claw\u002Fmicroclaw --skill dna-paint","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw\u002Fblob\u002Fmain\u002Fmicroclaw\u002Fskills\u002Fdna-paint\u002FSKILL.md",true,[],[46,53,59,65,71,77],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M72WAR49VE6T3ENRRZ523X76","extension-example","Exercise extension-dependent skill loading.","https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Fextension-example\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":19},"01M72WAPVCEKCRHRP3TNBCFNGF","weather","https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskill_packages\u002Fweather\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M72WANVZ372X54DH6P9AVXR4","metrics","https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskill_packages\u002Fmetrics\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M72WAM0BVCPRGQJ3QYZZYX7X","markdown","https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskill_packages\u002Fmarkdown\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M72WAJZAX9VVQA61FVPN5XVC","executable","https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskill_packages\u002Fexecutable\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M72WAH1XNYBSZFESQH6BTCJ2","conformance","https:\u002F\u002Fgithub.com\u002FMicro-Claw\u002Fmicroclaw\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskill_packages\u002Fconformance\u002FSKILL.md",[],[]]