forge-init
Skill4 starsChanged 3 months agoApache-2.0View on GitHub
Repo bootstrap runbook — writes the local agent contract and captures repo specifics. User-invoked only.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- fabioalencar/forgedesigntheir own version, first seen Oct 20, 2026
- tinyorbit-ai/skillstheir own version, first seen Aug 20, 2026
- tonmoy007/forge-pluginstheir own version, first seen Jun 9, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- tonmoy007/forge-plugins★ 2same name
- fabioalencar/forgedesign★ 1same name
- tinyorbit-ai/skills★ 1same name
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 repository35 other files it gives its agents.
- AGENTS.mdAGENTS.md
- 12-factor-appSkill
- caveman-commitSkill
- caveman-reviewSkill
- cavemanSkill
- conventional-commitsSkill
- creating-pull-requestsSkill
- deep-researchSkill
- dispatching-parallel-agentsSkill
- domain-modelingSkill
- finishing-a-development-branchSkill
- forge-cleanupSkill
- forge-docsSkill
- forge-eu-accessibilitySkill
- forge-flowSkill
- forge-seoSkill
- forge-setupSkill
- forgeSkill
- forge-tailwindcss-conventionsSkill
- frontend-designSkill
- git-issue-trackerSkill
- grillingSkill
- kiss-principleSkill
- loopsSkill
- planning-and-task-breakdownSkill
- prototypeSkill
- pr-resolveSkill
- pr-reviewSkill
- resolving-merge-conflictsSkill
- ste100Skill
- subagent-driven-developmentSkill
- use-git-identitySkill
- using-git-worktreesSkill
- verification-before-completionSkill
- wayfinderSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. 0 are common, 12 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
12
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 16
Read ./agents.template.md.
- Distinctive: line 19
Write the template content verbatim as AGENTS.md at repo root.
- Distinctive: line 24
Do not overwrite user-written sections.
- Distinctive: line 32
Never commit or stage AGENTS.md.
- Distinctive: line 40
After every grilling wave (4 questions answered), adapt AGENTS.md:
- Distinctive: line 43
Never add deployment-specific or coding-stack-specific content (frameworks, ORMs, runtimes, languages) — the agent reads the code for that.
- Distinctive: line 44
Ask the user: "Is this as intended?"
- Distinctive: line 45
Only add statements the you judge as unambiguous and valuable.
- Distinctive: line 49
Never emit code blocks. Never explain what code does.
- Distinctive: line 56
One-shot: Run once. Idempotent re-runs only update.
- Distinctive: line 60
Grilling is mandatory: Do not skip. User must confirm the contract.
- Distinctive: line 61
Goal-driven: Every addition to AGENTS.md states the goal, not the procedure.
- Distinctive: line 76
Create docs folders (forge-flow handles work surface)
“Common” is measured: kept by at least 6 in 10 of the 3 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
Forge Init
One-shot bootstrap. Runs once per repo. Prepares the agent's local contract (AGENTS.md) so every session starts aligned.
Steps
1. Create or smart-append AGENTS.md
Read ./agents.template.md. Use it as the base.
If AGENTS.md does not exist at repo root:
- Write the template content verbatim as
AGENTS.mdat repo root. - Load
use-git-identityto check ownership. If repo is under the configured identity, proceed. If unclear, ask: "Create AGENTS.md with forge contract at repo root?"
If AGENTS.md exists:
- Smart-append: preserve existing content, prepend the template's mandatory header (forge load rule + mandates rule) if absent, merge jargon entries (skip duplicates), keep letter-style prose.
- Do not overwrite user-written sections.
Always:
- Append at bottom (once):
--- Do not commit or stage this file. Local changes only. - Never commit or stage AGENTS.md.
Completion criterion: AGENTS.md at repo root contains the mandatory header, the jargon letter, the no-commit footer.
2. Grilling session for repo-specifics
Load grilling skill. Interview the user about this repo.
After every grilling wave (4 questions answered), adapt AGENTS.md:
- Amend only non-obvious facts not inferable from reading the repo.
- Interpret user's language into STE100.
- Never add deployment-specific or coding-stack-specific content (frameworks, ORMs, runtimes, languages) — the agent reads the code for that.
- Ask the user: "Is this as intended?"
- Only add statements the you judge as unambiguous and valuable.
During grilling:
- Answer user questions back in STE100.
- Never emit code blocks. Never explain what code does. Never document other files.
- Goal-driven prompting: frame every addition as "the goal is X, so the agent must Y" — no paths, no procedures, only the contract.
Completion criterion: grilling frontier empty; AGENTS.md captures repo-specific contract.
Rules
- One-shot. Run once. Idempotent re-runs only update.
- Never commit AGENTS.md. It is a local contract, not a repo artefact.
- STE100 throughout. Letter style; jargon: one flowing sentence per term.
- Mandatory header lives at top of AGENTS.md. Forge load rule + mandates rule. Never removed.
- Grilling is mandatory. Do not skip. User must confirm the contract.
- Goal-driven. Every addition to AGENTS.md states the goal, not the procedure.
- No stack-specific content. Frameworks, ORMs, runtimes, languages are inferable from the code — do not write them into AGENTS.md.
Template
The template lives at ./agents.template.md. Read it verbatim. It contains:
- Letter header — "This is a letter from me to you"
- Mandatory forge load rule
- Mandates rule
- Jargon — one flowing sentence per confusing term only
Boundaries
Forge Init does not:
- Create docs folders (forge-flow handles work surface)
- Set git identity (use-git-identity does that)
- Initialise labels (wayfinder/scripts/setup-repo-gh-cli.sh does that)
- Chart maps or resolve tickets (forge does that)
It only ensures AGENTS.md exists with the contract, then grills for repo specifics.
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.