[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7501535VCNP242JKSRRM2XD":3,"skill-same:01M7501535VCNP242JKSRRM2XD":53},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":20,"frontMatter":21,"refName":22,"commitSha":23,"skillDirectory":24,"forksCount":25,"repositoryHtmlUrl":26,"repositoryHomepageUrl":6,"ownerHtmlUrl":27,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":28,"install":29,"resources":42},"01M7501535VCNP242JKSRRM2XD","output-router",null,"hassan-mohiddin\u002Ffreeflow","hassan-mohiddin","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F63719517?v=4","01M75001RPX0N6YPMBKX9R29R3","Hassan Mohiddin","01M75001S94PQKYRZBT372B1DF",5,"MIT","https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002F.deprecated\u002Foutput-router\u002Fcapabilities\u002Foutput-router\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"# Output Router\n\nChoose how evidence moves into context.\n\nFreeflow tools are the safe first choice for unknown-size, exploratory, repo-wide, generated\u002Flog-adjacent, structured, or likely noisy output, including broad `rg`\u002F`find`\u002F`git`\u002Fhelp\u002Fdocs scans. Native tools stay direct for known-small, exact, intentionally raw, or mutating work. A post-hoc cap such as `head`, `sed -n`, `tail`, or `wc` does not make a broad producer known-small.\n\nOutput Router does not classify the task. Workflow, Decision Gate, or Discuss decides whether to answer, ask, explore, plan, or stop. Output Router starts after that route is clear.\n\n## First Decision\n\nAsk what kind of data you need:\n\n- Existing repo, explicit local, or vault data: use `freeflow_search`.\n- New command or script-produced output: use `freeflow_run`.\n- Several independent Freeflow operations: use `freeflow_batch`.\n- Config\u002Fstatus\u002Fvault\u002Fscript-adapter facts: use `freeflow_status`.\n- Enabled MCP\u002Fweb\u002Ffetch\u002Fcode-search output: call the host tool directly; Pi observed routing runs after the host result when configured.\n- Known whole file, intentionally exact shell behavior, cmux\u002FPi control probes, explicit fallback after a routed-source limitation, or edits: use native `read`, `bash`, `edit`, or `write`.\n\nWhen unsure how much output will return, use Freeflow first. Bound the producer before using native tools; do not rely on trimming the output after a broad command runs.\n\n## `freeflow_search`: Existing Data\n\nUse `freeflow_search` for repo files, explicit external local files, and vaulted output that already exists.\n\nActions:\n\n- `locate`: find likely paths\u002Foutput ids first. Use before reading when you only need candidates.\n- `query`: return the best focused evidence packets for a search question.\n- `get`: user gave exact-ish text\u002Fcode\u002Fsnippet or asks “where is this?” Find the best match and return source coordinates plus matched content.\n- `retrieve`: you already know the path\u002FoutputId and line range. Return those exact lines. Do not use this as search.\n- `expand`: widen a previous evidence packet.\n- `explain`: explain a previous routed decision or vault output\u002Frecoverability.\n- `transform`: compute facts or deterministic subsets from repo\u002Flocal\u002Fvault\u002Ffile\u002Foutput data.\n\nUse `source.kind=\"repo\"` for the current checkout. Use `source.kind=\"local\"` for external local docs\u002Ffiles with an explicit absolute `root` and optional relative `path`; do not put local roots in shared config. Use `source.kind=\"vault\"` for previous Freeflow outputs.\n\nVault search patterns:\n\n- Unknown output id: `source.kind=\"vault\"` without `outputId`, plus `query` and optional `filters`.\n- Known output id and unknown line: `action=\"query\"` or `get` with `source.outputId`.\n- Known output id and known line range: `action=\"retrieve\"` with `lineRange`.\n- Metadata-only result: use `action=\"explain\"`; do not claim raw recovery.\n\nUse `preserve=\"full\"` only when exact fidelity matters. It still means exact-under-caps, not unlimited context injection.\n\n## `freeflow_search action=transform`: Processing Existing Data\n\nUse transform when raw data is too large\u002Fstructured and you need computed facts.\n\nGood transform cases:\n\n- logs: status counts, error rates, slow examples,\n- JSON\u002FCSV: counts, groupings, top values, extracted fields,\n- test\u002Fbuild\u002Ftypecheck output already in the vault,\n- MCP\u002Fweb\u002Ffetch\u002Fcode-search payloads already routed into the vault,\n- deterministic regex filtering, match counts, grouping, dedupe, topN, URL\u002Fcitation extraction, line\u002Fsize stats.\n\nPrefer deterministic operations when enough:\n\n- `regexFilter`, `countMatches`, `jsonExtract`, `groupByRegex`, `dedupe`, `topN`, `extractUrls`, `extractCitations`, `lineStats`, `sizeStats`.\n\nUse scripts only when deterministic operations cannot express the fact:\n\n- `operation.kind=\"script\"` runs a sandboxed transform over vault sources.\n- `script` without `operation` runs the processing-engine script path over repo\u002Flocal\u002Fvault sources.\n- Sandboxed scripts require `scriptTransform.enabled=true` and a proof-backed adapter.\n- Raw script code is not persisted; metadata stores hashes\u002Flabels\u002Flimits.\n- No sandboxed script path may silently fall back to host shell execution.\n\nUnsafe unsandboxed processing is local-only, explicit per call, and only through the processing script policy. It must say `unsafe\u002Funsandboxed`. Never put that opt-in in shared config.\n\n## `freeflow_run`: New Output\n\nUse `freeflow_run` when the producer creates new output and routed evidence is enough.\n\nBase producers are mutually exclusive:\n\n- `command`: shell command through the host-approved runner.\n- `script`: sandboxed code-as-producer with no repo\u002Fhome\u002Fenv\u002Fnetwork\u002Fvault access.\n\nUse `command` for:\n\n- tests, builds, lint, typecheck, CI-like checks,\n- broad shell search with unknown output size,\n- commands likely to print logs, JSON, status, or diagnostics,\n- commands whose exact raw output may need recovery.\n\nUse `script` for:\n\n- generating data with code when no repo\u002Fhome\u002Fenv\u002Fnetwork access is needed,\n- safe code-as-producer execution through the same sandbox engine as transforms,\n- stdout\u002Fstderr capture that should route like command output.\n\nDo not use `freeflow_run script` when the code needs repo files, environment variables, network, package imports, or host filesystem access. Use a host command intentionally, or transform an explicitly captured source.\n\nRun modifiers:\n\n- `goal`: tell the router why the run exists, e.g. `verification`, `diagnosis`, `build`, `log analysis`.\n- `filters`: deterministic line filtering over captured stdout\u002Fstderr\u002Fcombined.\n- `scriptFilter`: sandboxed script over already captured stdout\u002Fstderr\u002Fcombined. It never reruns the base producer.\n- `preserve=\"full\"`: request exact output under caps with vault recovery over cap.\n\nRun storage:\n\n- Default `hybrid-dedupe` stores exact output when exactness-sensitive and metadata-only for small non-sensitive successes.\n- Failures, verification\u002Ftest\u002Fbuild\u002Ftypecheck\u002Fdiagnosis output, `preserve=\"full\"`, filters, script filters, reducers, script producers, and large\u002Fnoisy output are exactness-sensitive.\n- Exact duplicates may store current metadata pointing to a previous exact `outputId`.\n- Plain metadata-only records get rerun guidance, not exact recovery claims.\n\n## `freeflow_batch`: Independent Freeflow Work\n\nUse `freeflow_batch` when several Freeflow-owned `run` or `search` steps can happen independently.\n\nGood batch cases:\n\n- run several bounded searches in parallel,\n- run independent checks and ask `queries[]` to extract facts from child evidence,\n- gather multiple repo\u002Flocal\u002Fvault facts without injecting every child result.\n\nDo not use batch for:\n\n- sequenced workflows,\n- arbitrary external tool orchestration,\n- mutating work,\n- steps where one child needs another child’s output.\n\nChild outputs remain in `details.result.steps`; exact child run output remains recoverable by child `outputId`.\n\n## Native Tools\n\nUse native tools when direct behavior is the point:\n\n- `read`: known whole file or exact direct file content is intended.\n- `bash`: expected-small exact shell behavior, host access, cmux\u002FPi control probes, or bounded producer behavior is intentionally needed.\n- `edit` \u002F `write`: mutations.\n\nNative `bash`\u002F`read` remain valid for known-small exact checks, cmux\u002FPi control probes, exact file reads, or explicit fallback after a routed-source limitation; name the limitation and keep the fallback narrow.\n\nDo not run broad native commands just to see what comes back. `head`, `sed -n`, `tail`, `wc`, or similar output caps bound visible text, not producer scope or relevance. Likely-large native commands include repo-wide `rg`, `grep -R`, `find`, package scans, generated-artifact scans, docs\u002Fhelp scans, logs, broad `git diff\u002Flog`, full tests\u002Fbuilds\u002Flints\u002Ftypechecks, and one-off scripts with unknown output size.\n\nIf native read\u002Fbash safety net is enabled, large\u002Fnoisy native output may be replaced with labeled Freeflow-routed output. It must include an `outputId` and recovery instructions. If safety-net routing fails, native output passes through with a warning.\n\n## Observed Routing\n\nFor enabled Pi MCP\u002Fweb\u002Ffetch\u002Fcode-search producers, call the host tool directly. Freeflow observes the completed result after host execution.\n\nObserved routing:\n\n- is off by default,\n- is explicit per producer\u002Fserver,\n- does not grant permissions or block host tools,\n- can persist exact output, metadata-only, or nothing depending on config,\n- fails open with original host output if routing fails.\n\nMutating MCP\u002Ftool calls still require normal workflow\u002Finterview judgment before the host call. Observed routing is not permission approval.\n\n## Hard Rules\n\n- Do not pretend routed output is native output.\n- Do not silently shorten native output.\n- Capture raw output before transforming it.\n- Keep raw source recovery separate from transformed\u002Freduced output recovery.\n- Treat `preserve=\"full\"` as exact fidelity under caps, not unlimited context injection.\n- Preserve exact failure and verification evidence lines.\n- Keep `toolStatus`, `execution.status`, and `routing.status` separate.\n- Metadata-only records must never claim exact raw recovery.\n- Freeflow mode changes guidance strength only. It must not enable `postToolRouting`.\n- Do not write removed `capture` or `providers` config.\n- Do not offer `redacted`; it is future-only.\n\n## Recovery Pattern\n\nFor routed run\u002Fnative\u002Fobserved\u002Ftransform output, recover exact text with:\n\n```json\n{\n  \"action\": \"retrieve\",\n  \"source\": { \"kind\": \"vault\", \"outputId\": \"ffout_...\", \"stream\": \"stdout|stderr|combined|raw\" },\n  \"lineRange\": { \"start\": 1, \"end\": 20 },\n  \"preserve\": \"full\"\n}\n```\n\nWhen line numbers are unknown, use `query` or `get` first. Use `expand` when a returned evidence packet is too narrow. Use `explain` when recoverability is unclear.\n\n## Config\n\nUse `freeflow_status` to inspect effective config, vault writability\u002Findex state, script adapters, observed routing, and migration recommendations.\n\nUse Pi's `\u002Foutput-router enable` or repository settings surface for initial enablement. Read [setup guidance](references\u002Fsetup.md) before repairing capability config or changing optional routing features. After valid setup, Output Router owns explicit operational tuning requested through its capability surface.\n\nPersist operational config only after an explicit tuning request or a previously accepted setup branch:\n\n- `outputRouter`: `enabled`, `profile`, `postToolRouting`, `storagePolicy`, thresholds, vault root\u002Fretention, generated paths, noisy command hints.\n- `outputRouter.observedRouting`: explicit MCP servers and web\u002Ffetch\u002FcodeSearch producer persistence.\n- `outputRouter.scriptTransform`: enabled flag, languages, sandbox\u002Fnetwork\u002Flimits\u002Fraw-script persistence.\n\nMinimal setup stays only `defaultMode`. Missing optional sections mean built-in defaults.\n\nRead [the safety policy](references\u002Fsafety-policy.md) before changing routing policy, reviewing router behavior, or handling exactness-sensitive output.\n",{},"main","eda09a1c9c9e8f9119c705174eb219ea41d274d8",".deprecated\u002Foutput-router\u002Fcapabilities\u002Foutput-router",0,"https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow","https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin","2026-11-15 01:37:40",{"source":7,"skillName":5,"commands":30,"prompt":40,"zipAvailable":41},[31,34,37],{"tool":32,"command":33},"npx","npx skills add hassan-mohiddin\u002Ffreeflow --skill output-router",{"tool":35,"command":36},"bunx","bunx skills add hassan-mohiddin\u002Ffreeflow --skill output-router",{"tool":38,"command":39},"pnpm","pnpm dlx skills add hassan-mohiddin\u002Ffreeflow --skill output-router","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002F.deprecated\u002Foutput-router\u002Fcapabilities\u002Foutput-router\u002FSKILL.md",true,[43,49],{"path":44,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":47,"sourceHtmlUrl":48},"references\u002Fsafety-policy.md","REFERENCE","参考文档",2315,"https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002F.deprecated\u002Foutput-router\u002Fcapabilities\u002Foutput-router\u002Freferences\u002Fsafety-policy.md",{"path":50,"resourceType":45,"resourceTypeDesc":46,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fsetup.md",8081,"https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002F.deprecated\u002Foutput-router\u002Fcapabilities\u002Foutput-router\u002Freferences\u002Fsetup.md",[54,61,68,75,82,89],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":18},"01M750797ZRKMXW0S0VJJZN8E7","write-skill","Use when creating, revising, merging or splitting a skill or any other text an agent acts on: a SKILL.md and its description, the references and scripts in a skill, AGENTS.md, CLAUDE.md, system or runtime prompts, tool descriptions, subagent briefs or judge prompts. Also use when deciding where an instruction should live, and when a skill is read at the wrong time, not read at all, or read but not followed.","https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002Fskills\u002Fwrite-skill\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":18},"01M7506Z3TX46QMAJKS7Y9NKJN","verify-work","Use whenever you need to know whether something works or a claim is true: running tests, builds, linters or type checks; reproducing a bug or checking a fix; trying the real path in the running software, a browser or a CLI; measuring performance; and reading what the output actually shows, including failures and conflicting results. Read it before saying a change works, is fixed or is done, and to check your own change or document before reporting it done. Also use when writing, designing or changing a test, fixture or regression check; when a claim depends on an integration, host lifecycle, an installed package, rendered UI or browser behavior, or performance; and when setting up, following or updating a project's saved steps for launching its software and trying each feature as a user would, or rechecking them after the software changed.","https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002Fskills\u002Fverify-work\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":18},"01M7506N2HWH8F76NQ2RM74ZE3","setup-freeflow","Use when installing, activating or turning off Freeflow in a repository, changing or repairing its repository or personal configuration (.freeflow\u002Fconfig.json, .freeflow\u002Flocal.json), setting up the Freeflow store that keeps task memory (local only, or local + cloud with a private repository, in a cloud session too) and moving older task records into it, or checking whether the current host is loading Freeflow's guidance and skills.","https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002Fskills\u002Fsetup-freeflow\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":18},"01M7506A30JCAMVJN1ZC646NJ1","research-work","Use when a question has to be answered from evidence gathered and checked first: an in-depth report on a codebase, library, agent harness or tool; a survey, literature or systematic review of papers, docs or prior work; finding out what sources say before relying on them, and rechecking a published study's numbers; a study or experiment whose plan is fixed before the data, or a quick check on about twenty tasks of whether a change helped; writing a paper from results with its figures, references and PDF, and answering its reviewers. Also use when citing code at a pinned commit, deciding how strong a claim's evidence is, checking that each number traces to its log and each citation supports its claim, designing a measurement or a model judge, and computing statistics for a result (intervals, paired comparisons, clustered runs, rater agreement, pass@k, multiple comparisons, power, meta-analysis). Comparing options from facts already in hand is Discuss Work's; checking whether a change works is Verify Work's.","https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002Fskills\u002Fresearch-work\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":18},"01M75064ASQQPEVVDHY23S1HKX","execute-work","Use for any task that uses tools: reading or searching a repository, running commands, diagnosing, and any work that changes code, files or settings: implementing a feature, fixing a bug, adding or changing behavior, configuring, prototyping, refactoring, simplifying or cleaning up code, deleting dead code, or maintaining a project. Also use when something fails, breaks, regresses or slows down and the cause is unknown, or a fix did not hold; when migrating users, data, a schema, a service or traffic to a replacement and removing the old path; when adding edge-case handling or code comments; and when handing work to subagents or parallel workers, or working from another agent's brief. Read it before your first tool call on a task, even a small or read-only one: it reads the request for its asks and implied requirements, finds what the change touches before you edit, and reports the result.","https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002Fskills\u002Fexecute-work\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":18},"01M7505YQMR8ZX4X942R789XAM","evaluate-behavior","Use when testing, evaluating or comparing how an agent behaves: whether a skill triggers, whether a change to a prompt, skill, model, setting or extension changes outcomes, measuring a reviewer or checker, or benchmarking agents on coding tasks, conversations or pipelines. Also use when writing or checking an eval's JSON files, running an eval and reading its results, and deciding what a result can claim, how many cases and trials it needs and what it costs.","https:\u002F\u002Fgithub.com\u002Fhassan-mohiddin\u002Ffreeflow\u002Fblob\u002Fmain\u002Fskills\u002Fevaluate-behavior\u002FSKILL.md",[],[]]