gather-context
Skill0 starsChanged 25 days agoMITView on GitHub
Read and understand requirements from any source (Jira/Linear/GitHub ticket, Slack thread, Notion/ Confluence/Google doc, PR, error traceback, pasted text), trace the code they touch, and list acceptance criteria, blockers and open questions. Used first by implement, spec, investigate, pr-review and address-comments; also on "read this ticket", "what does this ticket need", "understand the requirements", "map blockers".
2 other projects use a version of this.
Adopted by
Most recently taken up by
- dmzoneill/skillstheir own version, first seen Oct 24, 2026
- t1djani/servotheir own version, first seen Jun 13, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- dmzoneill/skills★ 2same name
- t1djani/servo★ 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 repository18 other files it gives its agents.
- address-commentsSkill
- demoSkill
- deploySkill
- documentSkill
- implementSkill
- investigateSkill
- observeSkill
- pr-follow-upSkill
- project-profileSkill
- pr-reviewSkill
- qa-planSkill
- qaSkill
- resolve-conflictsSkill
- specSkill
- sprintSkill
- standupSkill
- todoSkill
- triageSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Follow ../../shared/policy.md.
- Distinctive: line 21
Follow links one level deep when they look like requirements.
- Distinctive: line 22
Related work: search the tracker and PR host for the ticket key and key terms: earlier PRs, duplicate or sibling tickets.
- Very specific: line 30
Read the repo's CLAUDE.md/README for conventions.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Gather context
Follow ../../shared/policy.md. Run dev-skills:project-profile if not loaded this session.
Reads only.
1. Fetch every input
Identify each input's type from its shape (ticket key/URL, Slack link, doc URL, PR URL, stack trace).
- Ticket: per
../../shared/references/trackers.md. Description and comments (requirements often live only in comments), linked issues, parent epic, attachments list, linked docs/designs. - Docs/threads: Notion, Confluence, Google Docs, Slack through connected tools (ToolSearch). Follow links one level deep when they look like requirements.
- Related work: search the tracker and PR host for the ticket key and key terms: earlier PRs, duplicate or sibling tickets.
- Unreachable source: note it as a gap; continue with the rest.
- Fetch inside one
Agentcall,model: haiku(policy "Delegating big reads"), returning verbatim: description, comments (author, date), linked issues and docs, attachment names, related PRs and tickets. Keep steps 2 and 3 in the main thread.
2. Trace the code
Read the repo's CLAUDE.md/README for conventions. Find the entry points the requirement touches
and trace end to end (route → logic → data → jobs/webhooks → frontend if relevant). Note file:line
for each place that will change or constrain the change. Check git log on those files for recent
related changes.
3. Find the gaps
- Acceptance criteria: explicit ones verbatim; inferred ones marked
(inferred). - Unknowns: anything the inputs and code cannot answer. For each: the question, why it matters, and the assumption to proceed with.
- Blockers: things that stop the work (missing access, dependency on another team's change, contradictory requirements).
- Resolve what you can from code, other tickets, and docs before calling it unknown.
Apply the mode (policy "Autonomy"): in guided ask the unknowns now; in balanced ask only an
unknown that would change the work materially; otherwise keep the assumptions.
Output
Write ~/.dev-skills/work/<project-slug>/<TICKET or short-slug>/context.md:
# <TICKET> — <title>
Sources: <links read> · Gaps: <sources not reachable>
## Summary
<3-5 lines: what is asked and why>
## Acceptance criteria
- [ ] ...
## Code map
- `path:line` — <role in the change>
## Related
- <PR/ticket> — <relation>
## Unknowns
1. <question> — matters because <...> — assuming <...>
## Blockers
- <blocker or "none">Return the path and the summary. Questions for product or the manager are drafts: they go to the confirmation list only when the user wants them sent.
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.