init
Skill5 starsChanged 53 days agoMITView on GitHub
Use to configure managed apps and work routes, sync their instruction summaries, or check dispatch readiness.
137 other projects use a version of this.
Adopted by
Most recently taken up by
- naderelewa/Product-to-Prodtheir own version, first seen Nov 8, 2026
- ShahriarBijoy/show-me-howtheir own version, first seen Nov 8, 2026
- MMACABRE/SKILLStheir own version, first seen Nov 8, 2026
- BSSE23047/nauraiz-devtheir own version, first seen Nov 8, 2026
- edwinidrus/dragonhoardtheir own version, first seen Nov 8, 2026
- NilsHellwig/acl-paper-skillstheir own version, first seen Nov 8, 2026
Used with
Projects with it also use daily-arxiv, exp-eval and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
LambenthanSkilldaily-arxiv
每日从 arXiv 拉取新论文,过滤相关性,auto-ingest 高优先级论文,检测 SOTA 更新
Also in 5 projects that use this
LambenthanSkillexp-eval
实验判决门:Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges
Also in 5 projects that use this
skyllwtSkillexp-pilot-eval
Pilot result evaluation — read pilot results, apply success criteria, update idea page (pilot_result, failure_reason if failed), generate PILOT_VERDICT_REPORT. Called by /ideate Phase 5 after /exp-pilot-run.
Also in 4 projects that use this
skyllwtSkillexp-pilot-run
Pilot experiment execution — read Pilot Spec YAML, write pilot code, run experiment(Confirm with the user before operation and require the applicant to conduct manual inspection), return results. Called by /ideate Phase 5. Does NOT modify wiki pages or judge pass/fail.
Also in 4 projects that use this
gaebalaiSkillcodex-system
PROACTIVELY consult Codex CLI, your highly capable supporter with exceptional reasoning and task completion abilities. Codex is a trusted expert you should ALWAYS consult BEFORE making decisions on: design choices, implementation approaches, debugging strategies, refactoring plans, or any non-trivial problem. When uncertain, consult Codex. Don't hesitate - Codex provides better analysis. Explicit triggers: "think deeper", "analyze", "second opinion", "consult codex".
Also in 3 projects that use this
- +6 more from Lambenthan
- +1 more from skyllwt
Other versions137 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 137
- microsoft/vscode★ 195ksame name
- mcollina/skills★ 2ksame name
- skyllwt/OmegaWiki★ 1.7ksame name
- DAIR-AutoResearch/AutoSci★ 1.7ksame name
- PKU-DAIR/AutoSci★ 1.7ksame name
- skyllwt/AutoSci★ 1.7ksame name
- TokenRollAI/cc-plugin★ 483same name
- pipecat-ai/gradient-bang★ 420same name
- AlexPEClub/ai-coding-starter-kit★ 382same name
- dannysmith/tauri-template★ 312same name
127 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository18 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- agent-feedbackSkill
- asking-peer-agentsSkill
- boss-assistantSkill
- boss-saySkill
- bringing-coworkerSkill
- choosing-graphSkill
- contacting-orchestratorsSkill
- create-great-harnessSkill
- dispatching-workSkill
- handoff-orchestratorSkill
- i-am-orchestratorSkill
- notifying-main-agentSkill
- peeking-workSkill
- reporting-to-userSkill
- resolving-appSkill
- shipping-taskSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
14 rules. 0 are common, 6 are distinctive, and 8 are very specific to this project.
0
Common
Most of the 125 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
Follow the schema's legacy migration rules when the old path is the source.
- Distinctive: line 17
Use a separate app-rooted workroom when ownership or continuity warrants it.
- Very specific: line 24
Ask only about choices the request and evidence leave unresolved.
- Very specific: line 25
Write the resolved .straw-boss/apps.json, read it back through the shared handler, and preserve unrelated fields.
- Distinctive: line 29
Read existing routing sections in root AGENTS.md and CLAUDE.md.
- Very specific: line 40
Put confirmed routes between <!-- straw-boss:agent-routing:start/end --> markers in both root instruction files.
- Very specific: line 45
Before any dispatch, check the CLI/service (command -v herdr, herdr status), current pane and provider identity, and provider integration …
- Very specific: line 49
Create dispatch/ and dispatch/archive/ under Path.home() / ".straw-boss" when needed.
- Very specific: line 51
Every launch rechecks readiness through dispatching-work.
- Very specific: line 55
Check whether each relevant app has AGENTS.md and CLAUDE.md, reusing a current inventory.
- Very specific: line 60
Use the same inline/workroom choice as discovery.
- Distinctive: line 66
Create missing root instruction files with a project heading.
- Distinctive: line 67
Update only the managed-apps block in each file:
- Distinctive: line 80
Read back both files: app summaries must match the configuration, routing blocks must match the confirmed routes, and surrounding content …
“Common” is measured: kept by at least 6 in 10 of the 125 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Read existing configuration
Resolve the git root and use the shared read handler. Reuse confirmed app and routing choices. An existing configuration is the base for requested changes; an unchanged app needs only missing information checked. Follow the schema's legacy migration rules when the old path is the source.
Resolve apps and missing fields
Identify candidates from manifests and common app directories, then resolve any ambiguous target with the user. For bounded discovery, load the target app's instructions and inspect it directly. Use a separate app-rooted workroom when ownership or continuity warrants it; follow dispatch readiness before launching one through dispatching-work.
Propose fields from app evidence using the field reference.
Required fields are name, dir, and match.
Optional fields need evidence or user-supplied policy: redirects, lifecycle restrictions, provider defaults, git workflow skills, local files, and cross-app skill pointers.
localFiles lists existing untracked, gitignored files; use optional: true only when the app remains operable without the file, and identify sensitive entries.
Ask only about choices the request and evidence leave unresolved.
Write the resolved .straw-boss/apps.json, read it back through the shared handler, and preserve unrelated fields.
Configure work routes
Read existing routing sections in root AGENTS.md and CLAUDE.md.
Reuse an agreed section; resolve conflicting values with the user.
Configure routes when requested or when dispatch needs an unresolved setup.
For a new or changed route, recommend a complete setup from local provider configuration and the user's model preferences; consult current official guidance when local evidence is insufficient. Each route records:
<work description> → worker: kind=<kind>, profile=<profile|default>, model=<model|default>, effort=<effort|default>; advisor=<model|none>
Confirm unresolved choices for that route as a whole.
A Claude route may use a supported native advisor; Codex records advisor=none.
Put confirmed routes between <!-- straw-boss:agent-routing:start/end --> markers in both root instruction files.
Preserve text outside the markers.
Check dispatch readiness
Before any dispatch, check the CLI/service (command -v herdr, herdr status), current pane and provider identity, and provider integration (herdr integration status).
A missing Claude integration can be installed with herdr integration install claude under the user's authorization for its global hook/settings change.
Codex readiness uses its live provider session and terminal identity.
Create dispatch/ and dispatch/archive/ under Path.home() / ".straw-boss" when needed.
Report unmet dependencies separately from configuration completion; local configuration and instruction sync can finish without Herdr.
Every launch rechecks readiness through dispatching-work.
Fill missing instructions
Check whether each relevant app has AGENTS.md and CLAUDE.md, reusing a current inventory.
Existing .agents/skills/, .claude/skills/, rules, and hooks also supply app-owned instructions.
Offer create-great-harness for missing files while preserving existing instructions.
Carry an existing bootstrap authorization; ask only for an unrequested addition.
Use the same inline/workroom choice as discovery.
Source changes follow the target's leveraging-tasks workflow when available.
Sync root summaries
Finish any repo-root bootstrap before editing the same files here. Create missing root instruction files with a project heading. Update only the managed-apps block in each file:
<!-- straw-boss:apps:start -->
## Managed apps (straw-boss)
api — apps/api
web — apps/web
Full config: `.straw-boss/apps.json`.
<!-- straw-boss:apps:end -->Read back both files: app summaries must match the configuration, routing blocks must match the confirmed routes, and surrounding content must be preserved.
Complete when: requested configuration and instruction sync are verified, dispatch readiness is reported, and any bootstrap is complete or has a recorded next event.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

