spec-kit
Skill0 starsChanged 26 days agoMITView on GitHub
Help adopt GitHub Spec Kit for Spec-Driven Development (SDD), initialize it for an AI agent, and hand its feature artifacts to implement. Use when users mention GitHub Spec Kit, specify CLI, /speckit-* skills, or setting up SDD in a repository.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- aaronpdoherty/claude-skillstheir own version, first seen Jul 18, 2026
- rloisell/rl-agents-n-skillstheir own version, first seen Jun 5, 2026
- shaunburdick/skillstheir own version, first seen Apr 19, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- rloisell/rl-agents-n-skills★ 8same name
- shaunburdick/skills★ 1same name
- aaronpdoherty/claude-skills★ 0same 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 repository10 other files it gives its agents.
- .github/copilot-instructions.mdCopilot instructions
- ado-implementSkill
- branch-policiesSkill
- grillingSkill
- grill-meSkill
- implementSkill
- lintSkill
- pr-reviewSkill
- prSkill
- teachSkill
The file
We split this skill into the rules it gives an agent
12 rules. 0 are common, 5 are distinctive, and 7 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
7
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Use the current official docs and the installed CLI's --help output for command syntax.
- Distinctive: line 14
Keep all initialization and generated artifacts in the user's target project, not in this plugin.
- Very specific: line 20
If specify is missing, explain that the official CLI is separately installed, for example with uv tool install specify-cli.
- Very specific: line 21
Ask which supported agent integration to initialize.
- Very specific: line 22
Before initialization, show the exact command and target directory.
- Distinctive: line 23
After initialization, inspect the CLI output and generated file list.
- Very specific: line 40
When the user wants the feature delivered as a tested pull request, offer to hand the active Spec Kit feature to implement:
- Distinctive: line 42
Ask for the active feature directory or the specific Spec Kit artifacts to use.
- Distinctive: line 43
Read the selected feature's specification, plan, task list, and any relevant project constitution, using the paths and artifact names …
- Very specific: line 45
Start or continue implement with the selected artifacts as its local-file/task source.
- Very specific: line 46
Do not mark Spec Kit tasks complete based only on checkboxes.
- Distinctive: line 48
If the user only asks a question about SDD or Spec Kit, answer conversationally with official links.
“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
Help the user adopt GitHub's Spec Kit in their project. This skill is an onboarding and handoff layer; it does not bundle, install, or replace the Spec Kit CLI, generated skills, templates, or workflows.
Ownership and sources
- Spec Kit is maintained at github/spec-kit and its authoritative documentation is github.github.io/spec-kit.
specifygenerates and manages the project's Spec Kit files and/speckit-*skills. Do not copy, recreate, hand-edit, or claim ownership of those generated files.- Use the current official docs and the installed CLI's
--helpoutput for command syntax. Spec Kit evolves; do not assume an older command, artifact layout, or integration name is still current. - Keep all initialization and generated artifacts in the user's target project, not in this plugin.
Set up Spec Kit
- Confirm the target repository/workspace and whether the user wants to initialize Spec Kit there or only learn how an existing setup works. Do not assume the plugin repository is the target project.
- Inspect prerequisites using available read-only commands and the official installation guide. Spec Kit documents Python 3.11+ and
uvas prerequisites. Checkpython3 --version,uv --version, andspecify --help/specify versionif available; report missing tools explicitly. - If
specifyis missing, explain that the official CLI is separately installed, for example withuv tool install specify-cli. Show the install command and ask before running it. - Ask which supported agent integration to initialize. Recommend
copilotonly when the user is using GitHub Copilot; verify current integration keys and init options withspecify init --helpor official docs. - Before initialization, show the exact command and target directory.
specify initwrites project files, so ask for approval before running it. Never run it against a non-empty project without checking what it would create or overwrite. - After initialization, inspect the CLI output and generated file list. Report the generated skill names and how to invoke them, without modifying those files. If verification fails, report the error and stop rather than claiming setup succeeded.
Official starting points:
Explain the workflow
Describe the current workflow from the official reference instead of presenting a stale command list as authoritative. The core flow is generally to establish project principles, specify a feature, plan it, break it into tasks, implement, and converge; optional clarification, checklist, and analysis steps provide quality gates. Link the relevant official reference and let the CLI's installed version determine the actual skill names and syntax.
Spec Kit also has optional processes/extensions beyond feature development. Do not enable them unless the user asks and confirms the choice.
Hand off to implement
When the user wants the feature delivered as a tested pull request, offer to hand the active Spec Kit feature to implement:
- Ask for the active feature directory or the specific Spec Kit artifacts to use. Do not guess which feature is active or scan unrelated feature folders.
- Read the selected feature's specification, plan, task list, and any relevant project constitution, using the paths and artifact names present in that repository.
- Summarize the requested outcome and derive testable acceptance criteria. Treat the selected specification as task context, not as permission to change files or skip
implement's interview, controls, branch, test, review, and PR gates. - Start or continue
implementwith the selected artifacts as its local-file/task source. Preserve the source paths for traceability. Keep Spec Kit artifacts read-only unless the user explicitly requests their update. - Do not mark Spec Kit tasks complete based only on checkboxes;
implementmust verify every agreed criterion with evidence.
If the user only asks a question about SDD or Spec Kit, answer conversationally with official links. Do not initialize a project or create learning files unless asked.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.

