[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7B0J0X2Q8HMARJBCQGKGS3D":3,"skill-same:01M7B0J0X2Q8HMARJBCQGKGS3D":80},{"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":43,"commitSha":44,"skillDirectory":45,"forksCount":14,"repositoryHtmlUrl":46,"repositoryHomepageUrl":6,"ownerHtmlUrl":47,"ownerBlogUrl":6,"ownerDescription":48,"updatedAt":49,"install":50,"resources":63},"01M7B0J0X2Q8HMARJBCQGKGS3D","open-design-mcp-agent-workflow",null,"Build via Open Design MCP; drive agentic CLIs as sub-agents.","dbillion\u002Fhermes-skills-public","dbillion","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F49393855?v=4","01M7AVH8QPGCM8WKWS22TRT0T6","Oludayo Adeoye","01M7AVH8RD1JJWA0YS9CFEWAV9",0,"MIT","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fopen-design-mcp-agent-workflow\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: open-design-mcp-agent-workflow\ndescription: \"Build via Open Design MCP; drive agentic CLIs as sub-agents.\"\nversion: 1.0.0\nauthor: Hermes Agent (captured from user workflow)\nlicense: MIT\nplatforms: [linux, macos, windows]\ntags: [open-design, mcp, acp, design-automation, carousel, pdf, agent-adapter, subagent]\nrelated_skills: [build-discipline, writing-plans, verification-loop]\n---\n\n# Open Design MCP + Agent-CLI Workflow\n\nHow to use Open Design (github.com\u002Fnexu-io\u002Fopen-design) as a **toolbox** that Hermes\ndrives, and how to extend it with other agentic CLIs as sub-agents. Captured from a\nreal end-to-end build: a travel carousel → 8 PNG cards + PDF, plus enumerating\u002Fharvesting\na Telegram account's channels.\n\n## Mental model: two directions, only one we use\n\n- **MCP (Hermes → OD)** — Hermes reaches INTO OD and calls its tools. OD is the\n  toolbox; Hermes is the builder. **This is the path we use.** It works regardless of\n  OD's run\u002Fexecution state.\n- **Agent adapter \u002F ACP (OD → Hermes)** — OD dispatches tasks TO Hermes as an execution\n  engine. Useful only when *OD's* UI is driving. In practice the `start_run` execution\n  path was **broken** in our setup (returns `202` then `status: failed`), so we do NOT\n  rely on it. We build by reading OD skill assets directly via MCP and rendering locally.\n  Don't claim OD \"executes\" a skill for you — it didn't, and it's the reverse direction.\n\n**User's own framing (verbatim intent):** *\"we can just call its mcp to build, we are\nthe one building by using its mcps.\"* — the builder is Hermes; OD supplies skills\u002Ftemplates.\n\n## Daemon (the MCP server host)\n\n- OD runs a local privileged daemon. Keep it alive as a `systemd --user` service\n  (`open-design-daemon.service`), pinned to the Node version its native addons need\n  (Node 24.19.0 for `better-sqlite3` ABI 137). Auto-starts on boot.\n- Health: `GET http:\u002F\u002F127.0.0.1:7456\u002Fapi\u002Fhealth` → `200`. Also `\u002Fapi\u002Fskills`\n  (162 skills), `\u002Fapi\u002Fplugins` (150), `\u002Fapi\u002Fprojects`.\n- The agent adapter is registered (`hermes.ts` in the daemon's defs) and passes\n  `hermes acp --check`, but that does NOT mean OD can execute runs — see caveat above.\n\n## MCP tool surface (the real interface)\n\n- After a gateway restart that reloads MCP, the tools appear as `mcp_open_design_*`.\n  In a continuing session they may not be inline until a `\u002Freset` — but you can always\n  drive them directly via the daemon's stdio JSON-RPC (`node cli.js mcp --daemon-url\n  http:\u002F\u002F127.0.0.1:7456`), which returns **22 tools** including:\n  `list_skills`, `list_plugins`, `create_project`, `create_artifact`, `write_file`,\n  `get_file`, `start_run`, `get_run`, `list_agents`, `collect_brief`, `confirm_brief`.\n- Working build loop (proven): `create_project` → `write_file` (base64) ×N → files land\n  in OD's artifact\u002Fproject store. `start_run` is the broken execution path; don't use it.\n\n## Building a deliverable (proven: travel carousel)\n\n1. **Read the skill** via MCP `list_skills`; pick the template (e.g. `card-xiaohongshu`\n   for a swipeable carousel, `deck-swiss-international` for a deck, `brandkit` for brand\n   assets, `canvas-design` for PNG\u002FPDF). Skill assets live under `skills\u002F\u003Cid>\u002F` in the OD\n   repo; copy the HTML\u002FTailwind scaffold.\n2. **Author the content** (cards, copy, palette) as HTML\u002FCSS. Keep the template's\n   structure (vertical `.deck` of `.card` divs, 1080×1440 \u002F 3:4).\n3. **Media**: OD's `fal-*` image\u002Fvideo skills need a fal.ai key — which was **LOCKED\n   (HTTP 403, reason TOP_UP)** in our run. Fallback that worked: **Pexels** via API key\n   (`api.pexels.com\u002Fv1\u002Fvideos\u002Fsearch`) for stock video\u002Fpics. Download real clips, wire\n   `\u003Cvideo>`\u002F`\u003Cimg>` with `poster=` frames (videos don't autoplay in a static screenshot).\n4. **Render to upload-ready output** (local, not OD): headless Chromium screenshots each\n   card → PNG; bundle PNGs into a PDF with PIL. See `scripts\u002Frender_carousel.py`.\n5. **Write deliverables into OD** via MCP `write_file` so they show in the OD UI.\n\n## Rendering cards → PNG + PDF (the local step)\n\nOD skills produce HTML\u002FCSS, not image files. To get postable assets:\n- **Per-card HTML**: extract each `.card` div into its own minimal HTML file (the\n  `?card=N` query-param isolation trick FAILS under headless Chrome — it ignores the\n  param; you get 8 identical screenshots). Building one file per card is the reliable way.\n- **Screenshot**: `chromium --headless --no-sandbox --window-size=1080,1440\n  --screenshot=out.png file:\u002F\u002F\u002Fpath\u002Fcard-N.html`. Crop to 1080×1440 if the viewport\n  differs.\n- **PDF**: `PIL.Image.save(pdf, save_all=True, append_images=[...])` → one page per card.\n- **Video cards**: put a `poster=` still (extract with `ffmpeg -ss 2 -i clip.mp4 -frames:v 1`)\n  so the static card shows footage; the live `\u003Cvideo>` stays for the HTML deck.\n- Gotcha: a \"full page\" `--screenshot` only captures the first viewport, not the whole\n  scrollable deck — screenshot per card, never the whole page.\n\n## Driving other agentic CLIs as sub-agents\n\nVerified on the host (find via `command -v`): `pi`, `kilo`, `kiro`, `antigravity`,\n`cursor`, `gemini`, `codex`, `claude`, `qwen` are present; `opencode`\u002F`devin`\u002F`grok`\u002F\n`deepseek`\u002F`vela` are NOT. To test one as a sub-agent:\n- Run it non-interactively with a `-p`\u002F`exec` prompt; check auth. **Only `pi` returned a\n  clean authenticated answer** (`pi -p \"...\"` → works, has read\u002Fbash\u002Fedit\u002Fwrite tools).\n  `claude`\u002F`gemini` said \"not logged in\"; `codex` errored on skill YAML; `qwen` hung;\n  `antigravity` launched a browser and hung. So **`pi` is the one proven sub-agent**; the\n  rest need auth or are broken in this env. Don't claim the others work.\n- `pi` can read a repo, analyze code, and propose commands — it successfully diagnosed\n  tgforwarder's forward logic and produced the correct (after flag-fix) commands.\n\n## Pitfalls (this user's triggers — avoid)\n\n- **Claiming OD \"ran the skill\"** — it didn't via start_run; we built from its assets.\n- **Per-card screenshot via query param** — produces identical images; build one file per card.\n- **fal.ai assumed working** — it was LOCKED (TOP_UP). Use Pexels stock as the fallback.\n- **Claiming claude\u002Fcodex\u002Fgemini are usable sub-agents** — only `pi` was verified.\n- **\"Excuses \u002F it can't be done\"** — user mandate is *find a way*. Diagnose, then deliver.\n\n## Verification\n\n- Daemon `HTTP 200` on `\u002Fapi\u002Fhealth`.\n- MCP `tools\u002Flist` returns 22 tools.\n- Deliverable: open the HTML deck in a browser; confirm 8 distinct PNGs (different\n  file sizes = different content) + a multi-page PDF; visually check 1-2 cards.\n\n## Reference files (session-specific detail)\n- `references\u002Fpexels-fal-media-fallback.md` — exact Pexels search\u002Fdownload calls and the\n  fal.ai 403 TOP_UP finding, so a future session doesn't re-discover the lock.\n- `references\u002Fagent-cli-probe.md` — the non-interactive probe results for each CLI\n  (pi works; others need auth\u002Fbroken).\n- `scripts\u002Frender_carousel.py` — the Chromium+PIL render script (per-card HTML → PNG → PDF).\n","# Open Design MCP + Agent-CLI Workflow\n\nHow to use Open Design (github.com\u002Fnexu-io\u002Fopen-design) as a **toolbox** that Hermes\ndrives, and how to extend it with other agentic CLIs as sub-agents. Captured from a\nreal end-to-end build: a travel carousel → 8 PNG cards + PDF, plus enumerating\u002Fharvesting\na Telegram account's channels.\n\n## Mental model: two directions, only one we use\n\n- **MCP (Hermes → OD)** — Hermes reaches INTO OD and calls its tools. OD is the\n  toolbox; Hermes is the builder. **This is the path we use.** It works regardless of\n  OD's run\u002Fexecution state.\n- **Agent adapter \u002F ACP (OD → Hermes)** — OD dispatches tasks TO Hermes as an execution\n  engine. Useful only when *OD's* UI is driving. In practice the `start_run` execution\n  path was **broken** in our setup (returns `202` then `status: failed`), so we do NOT\n  rely on it. We build by reading OD skill assets directly via MCP and rendering locally.\n  Don't claim OD \"executes\" a skill for you — it didn't, and it's the reverse direction.\n\n**User's own framing (verbatim intent):** *\"we can just call its mcp to build, we are\nthe one building by using its mcps.\"* — the builder is Hermes; OD supplies skills\u002Ftemplates.\n\n## Daemon (the MCP server host)\n\n- OD runs a local privileged daemon. Keep it alive as a `systemd --user` service\n  (`open-design-daemon.service`), pinned to the Node version its native addons need\n  (Node 24.19.0 for `better-sqlite3` ABI 137). Auto-starts on boot.\n- Health: `GET http:\u002F\u002F127.0.0.1:7456\u002Fapi\u002Fhealth` → `200`. Also `\u002Fapi\u002Fskills`\n  (162 skills), `\u002Fapi\u002Fplugins` (150), `\u002Fapi\u002Fprojects`.\n- The agent adapter is registered (`hermes.ts` in the daemon's defs) and passes\n  `hermes acp --check`, but that does NOT mean OD can execute runs — see caveat above.\n\n## MCP tool surface (the real interface)\n\n- After a gateway restart that reloads MCP, the tools appear as `mcp_open_design_*`.\n  In a continuing session they may not be inline until a `\u002Freset` — but you can always\n  drive them directly via the daemon's stdio JSON-RPC (`node cli.js mcp --daemon-url\n  http:\u002F\u002F127.0.0.1:7456`), which returns **22 tools** including:\n  `list_skills`, `list_plugins`, `create_project`, `create_artifact`, `write_file`,\n  `get_file`, `start_run`, `get_run`, `list_agents`, `collect_brief`, `confirm_brief`.\n- Working build loop (proven): `create_project` → `write_file` (base64) ×N → files land\n  in OD's artifact\u002Fproject store. `start_run` is the broken execution path; don't use it.\n\n## Building a deliverable (proven: travel carousel)\n\n1. **Read the skill** via MCP `list_skills`; pick the template (e.g. `card-xiaohongshu`\n   for a swipeable carousel, `deck-swiss-international` for a deck, `brandkit` for brand\n   assets, `canvas-design` for PNG\u002FPDF). Skill assets live under `skills\u002F\u003Cid>\u002F` in the OD\n   repo; copy the HTML\u002FTailwind scaffold.\n2. **Author the content** (cards, copy, palette) as HTML\u002FCSS. Keep the template's\n   structure (vertical `.deck` of `.card` divs, 1080×1440 \u002F 3:4).\n3. **Media**: OD's `fal-*` image\u002Fvideo skills need a fal.ai key — which was **LOCKED\n   (HTTP 403, reason TOP_UP)** in our run. Fallback that worked: **Pexels** via API key\n   (`api.pexels.com\u002Fv1\u002Fvideos\u002Fsearch`) for stock video\u002Fpics. Download real clips, wire\n   `\u003Cvideo>`\u002F`\u003Cimg>` with `poster=` frames (videos don't autoplay in a static screenshot).\n4. **Render to upload-ready output** (local, not OD): headless Chromium screenshots each\n   card → PNG; bundle PNGs into a PDF with PIL. See `scripts\u002Frender_carousel.py`.\n5. **Write deliverables into OD** via MCP `write_file` so they show in the OD UI.\n\n## Rendering cards → PNG + PDF (the local step)\n\nOD skills produce HTML\u002FCSS, not image files. To get postable assets:\n- **Per-card HTML**: extract each `.card` div into its own minimal HTML file (the\n  `?card=N` query-param isolation trick FAILS under headless Chrome — it ignores the\n  param; you get 8 identical screenshots). Building one file per card is the reliable way.\n- **Screenshot**: `chromium --headless --no-sandbox --window-size=1080,1440\n  --screenshot=out.png file:\u002F\u002F\u002Fpath\u002Fcard-N.html`. Crop to 1080×1440 if the viewport\n  differs.\n- **PDF**: `PIL.Image.save(pdf, save_all=True, append_images=[...])` → one page per card.\n- **Video cards**: put a `poster=` still (extract with `ffmpeg -ss 2 -i clip.mp4 -frames:v 1`)\n  so the static card shows footage; the live `\u003Cvideo>` stays for the HTML deck.\n- Gotcha: a \"full page\" `--screenshot` only captures the first viewport, not the whole\n  scrollable deck — screenshot per card, never the whole page.\n\n## Driving other agentic CLIs as sub-agents\n\nVerified on the host (find via `command -v`): `pi`, `kilo`, `kiro`, `antigravity`,\n`cursor`, `gemini`, `codex`, `claude`, `qwen` are present; `opencode`\u002F`devin`\u002F`grok`\u002F\n`deepseek`\u002F`vela` are NOT. To test one as a sub-agent:\n- Run it non-interactively with a `-p`\u002F`exec` prompt; check auth. **Only `pi` returned a\n  clean authenticated answer** (`pi -p \"...\"` → works, has read\u002Fbash\u002Fedit\u002Fwrite tools).\n  `claude`\u002F`gemini` said \"not logged in\"; `codex` errored on skill YAML; `qwen` hung;\n  `antigravity` launched a browser and hung. So **`pi` is the one proven sub-agent**; the\n  rest need auth or are broken in this env. Don't claim the others work.\n- `pi` can read a repo, analyze code, and propose commands — it successfully diagnosed\n  tgforwarder's forward logic and produced the correct (after flag-fix) commands.\n\n## Pitfalls (this user's triggers — avoid)\n\n- **Claiming OD \"ran the skill\"** — it didn't via start_run; we built from its assets.\n- **Per-card screenshot via query param** — produces identical images; build one file per card.\n- **fal.ai assumed working** — it was LOCKED (TOP_UP). Use Pexels stock as the fallback.\n- **Claiming claude\u002Fcodex\u002Fgemini are usable sub-agents** — only `pi` was verified.\n- **\"Excuses \u002F it can't be done\"** — user mandate is *find a way*. Diagnose, then deliver.\n\n## Verification\n\n- Daemon `HTTP 200` on `\u002Fapi\u002Fhealth`.\n- MCP `tools\u002Flist` returns 22 tools.\n- Deliverable: open the HTML deck in a browser; confirm 8 distinct PNGs (different\n  file sizes = different content) + a multi-page PDF; visually check 1-2 cards.\n\n## Reference files (session-specific detail)\n- `references\u002Fpexels-fal-media-fallback.md` — exact Pexels search\u002Fdownload calls and the\n  fal.ai 403 TOP_UP finding, so a future session doesn't re-discover the lock.\n- `references\u002Fagent-cli-probe.md` — the non-interactive probe results for each CLI\n  (pi works; others need auth\u002Fbroken).\n- `scripts\u002Frender_carousel.py` — the Chromium+PIL render script (per-card HTML → PNG → PDF).\n",{"name":5,"tags":24,"author":33,"license":15,"version":34,"platforms":35,"description":7,"related_skills":39},[25,26,27,28,29,30,31,32],"open-design","mcp","acp","design-automation","carousel","pdf","agent-adapter","subagent","Hermes Agent (captured from user workflow)","1.0.0",[36,37,38],"linux","macos","windows",[40,41,42],"build-discipline","writing-plans","verification-loop","master","f341c46fed021656175d5c9ccc67a72126e50bfb","skills\u002Fopen-design-mcp-agent-workflow","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public","https:\u002F\u002Fgithub.com\u002Fdbillion","I enjoy writing codes,teaching programming, solving problems, tasted automation in Python and never looked back😋😋😋💞","2026-10-14 03:31:55",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add dbillion\u002Fhermes-skills-public --skill open-design-mcp-agent-workflow",{"tool":56,"command":57},"bunx","bunx skills add dbillion\u002Fhermes-skills-public --skill open-design-mcp-agent-workflow",{"tool":59,"command":60},"pnpm","pnpm dlx skills add dbillion\u002Fhermes-skills-public --skill open-design-mcp-agent-workflow","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fopen-design-mcp-agent-workflow\u002FSKILL.md",true,[64,70,74],{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"references\u002Fagent-cli-probe.md","REFERENCE","参考文档",1942,"https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fopen-design-mcp-agent-workflow\u002Freferences\u002Fagent-cli-probe.md",{"path":71,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":72,"sourceHtmlUrl":73},"references\u002Fpexels-fal-media-fallback.md",1851,"https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fopen-design-mcp-agent-workflow\u002Freferences\u002Fpexels-fal-media-fallback.md",{"path":75,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":78,"sourceHtmlUrl":79},"scripts\u002Frender_carousel.py","SCRIPT","脚本",3066,"https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fopen-design-mcp-agent-workflow\u002Fscripts\u002Frender_carousel.py",[81,88,95,104,111,118],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M7B25T0V7RBKES86Y3RTXQRN","blocked-page-recovery","Use when a fetch fails: 403\u002F429, paywall, WAF, bot wall.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fweb\u002Fblocked-page-recovery\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":19},"01M7B25DSBBMZRHVD28MAX7N5X","gemini-vision-fallback","Vision fallback: Gemini REST API if vision_analyze blocked.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fvision\u002Fgemini-vision-fallback\u002FSKILL.md",[],[],{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":99,"licenseFileUrl":100,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M7B22XW9QXT2ZWWXVZDFWYZP","twitter-cli","Use twitter-cli for ALL Twitter\u002FX operations — reading tweets, posting, replying, quoting, liking, retweeting, following, searching, user lookups. Invoke whenever user requests any Twitter interaction.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Ftwitter-cli\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Ftwitter-cli\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":19},"01M7B22QMWY1B4F7BC114RG4S6","prism-scan","Structural analysis through dynamically generated cognitive lenses. Generates the optimal analytical lens for the specific code\u002Fartifact, then executes it. Finds conservation laws, structural invariants, and concrete bugs that vanilla analysis misses. Use on any code file, system design, or text artifact.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fts-superhermes-prism-scan\u002FSKILL.md",[],[],{"skillNo":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":117,"publishedAt":19},"01M7B22JA5XCAP1ACT6V50WYR1","prism-reflect","Constraint transparency: analyzes an artifact structurally, then analyzes what its own analysis concealed. Produces a conservation law AND a constraint report showing what was maximized, what was sacrificed, and what to investigate next. The only AI skill that knows what it can't see.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fts-superhermes-prism-reflect\u002FSKILL.md",[],[],{"skillNo":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":122,"categories":123,"keywords":124,"publishedAt":19},"01M7B22EB15M75M4X3C52YGDCC","prism-full","Full Prism: multi-pass structural analysis with mandatory adversarial self-correction. Designs custom analytical passes, executes them with chaining, then attacks its own findings before synthesizing. Use for maximum depth on important code or artifacts.","https:\u002F\u002Fgithub.com\u002Fdbillion\u002Fhermes-skills-public\u002Fblob\u002Fmaster\u002Fskills\u002Fts-superhermes-prism-full\u002FSKILL.md",[],[]]