github-research
Skill0 starsChanged 3 months agoMITView on GitHub
Use when the user asks to search GitHub, find repositories, check what a project's code/issues/PRs/releases say, or look up how something is implemented in open source. Triggers on "search GitHub", "find a repo for", "check the issues on", "how do they implement X", "github research". Produces evidence records (URL pinned to a commit SHA + verbatim quote + date), never conclusions. Do NOT use for general web search (use web-research), Reddit/community discussion (use reddit-research), or when the user just wants you to write code (that's normal coding, not research).
10 other projects use a version of this.
Adopted by
Most recently taken up by
- EtanHey/golemstheir own version, first seen Oct 28, 2026
- block/buzztheir own version, first seen Oct 13, 2026
- willlee1995/skillstheir own version, first seen Oct 13, 2026
- maharshi-coding/oss-agenttheir own version, first seen Oct 11, 2026
- block/sprouttheir own version, first seen Oct 8, 2026
- shimenghan6/claude-code-startertheir own version, first seen Jul 22, 2026
Other versions10 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 10
- block/buzz★ 37ksame name
- block/sprout★ 37ksame name
- lingzhi227/agent-research-skills★ 350same name
- sergeeey/Claude-cod-top-2026★ 9same name
- dongzhigang13305312738-art/paper-skills★ 1same name
- shimenghan6/claude-code-starter★ 1same name
- srednoff888-art/srednoff-os-for-claude★ 1same name
- willlee1995/skills★ 1same name
- EtanHey/golems★ 0same name
- maharshi-coding/oss-agent★ 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.
- academic-researchSkill
- blog-researchSkill
- news-researchSkill
- official-source-researchSkill
- reddit-researchSkill
- research-intelligenceSkill
- research-plannerSkill
- social-researchSkill
- source-verifySkill
- web-researchSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 0 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 9 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 18
Don't use for: general web content (→ web-research), Reddit discussion even if it's about a GitHub project (→ reddit-research), …
- Very specific: line 22
Run gh auth status first. Authenticated unlocks code search (unavailable anonymously — GitHub returns 401 with no anonymous mode, not a …
- Very specific: line 35
If a search returns nothing: run references/query-relaxation.md, dropping the language filter first, then reducing to core terms, then a …
“Common” is measured: kept by at least 6 in 10 of the 9 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
GitHub Research
Overview
Collects evidence from GitHub: repos, code, issues, PRs, releases, discussions, commits. Each output is one evidence record — a claim backed by a verbatim quote pinned to a permanent anchor (commit SHA, not a branch that can move). This skill never concludes or synthesizes; it produces raw material for state/runs/<slug>/evidence.jsonl.
Layer rule: produces evidence, not results. Synthesis and confidence status are the orchestrator's job.
When to use
Direct use: "find repos that do X", "what does the changelog say about Y", "check if this bug is still open". As a component: orchestrator dispatches this alongside web/reddit research for implementation-type claims.
Don't use for: general web content (→ web-research), Reddit discussion even if it's about a GitHub project (→ reddit-research), writing/editing code in the current project (not a research task).
Capability check (once per session, not per query)
Run gh auth status first. Authenticated unlocks code search (unavailable anonymously — GitHub returns 401 with no anonymous mode, not a rate limit to route around) and higher limits. See references/tooling.md for the full chain and measured rate limits.
Workflow
- Pick the right search surface for the question:
- Repo discovery →
gh search repos(topic/stars/language filters) - "How is X implemented" →
gh search code(auth required; returnstextMatches— the verbatim quote, direct from the API) - "Is this a known issue" / demand signal →
gh search issuessorted by reactions, orgh api .../issues/N/comments - "What changed in version Y" →
gh release view/gh release list(publish date comes free) - Community discussion on the repo → GraphQL
search(type:DISCUSSION, ...)(not covered bygh search) Exact commands are in references/tooling.md — use them verbatim, don't improvise flag names.
- Repo discovery →
- Go past the README when the claim is "this is actually implemented/used". A repo existing, or its README claiming a feature, is not evidence the feature works. Follow the depth ladder: README → repo structure → the actual file implementing the claim → issues/PRs discussing it → releases confirming it shipped. Set
quality.repro.typehonestly:actual-runonly when the record shows a real command/code and its result;maintainer-confirmedwhen the author asserts it works without showing a run. Onlyactual-runsatisfies the implementation bar forconfirmed(engine/ranking.md) — "the maintainer said yes" is testimony, not reproduction. - Root follows who is speaking, not where the evidence lives. The project's own artifacts — README, docs, code, releases, commits, maintainer statements — are one project speaking regardless of which of these you quote:
root_id: github:owner/repo. But an outside, unaffiliated user's issue, comment, or reproduction is that person's voice, not the project's —root_id: person:<username>, per engine/origin-tracing.md. Getting this backwards in either direction breaks independence counting: collapsing an outside reporter intogithub:owner/repoerases real third-party corroboration (two unaffiliated users independently hitting the same bug are two roots, not one); conversely, assigning the maintainer's own release notes aperson:id manufactures false independence for a claim the project is making about itself. Decide by checking who wrote the specific artifact you're quoting — the repo owner/org, a listed maintainer, or an unrelated contributor/reporter — not by which repo it's filed under. - If a search returns nothing: run references/query-relaxation.md, dropping the language filter first, then reducing to core terms, then a single term. Log full exhaustion as a gap.
- Pin the quote correctly:
- Code: rewrite any
blob/main/...-style URL toblob/<full-sha>/...before recording it — branch-pinned URLs drift. - Comments: record
created_ataspublished_atANDupdated_atasmodified_at— an edited comment quoted from stale memory is a trap. - Code search hits carry no date; follow up with
gh api repos/.../commits?path=...&per_page=1for the commit date.
- Code: rewrite any
- Archive per references/citation-rules.md.
- Assign tier per references/source-tiers.md's github table (repo's own source/README/releases = tier 1; maintainer comments on their own issues = tier 3; drive-by comments/unmerged PRs = tier 4) and purpose.
- Write the evidence record per engine/evidence-contract.md,
verification_status: pending. Do not self-verify.
Output
One evidence record per claim, appended to the run's evidence.jsonl.
Common mistakes
| Mistake | Fix |
Citing a blob/main/... URL | Rewrite to the commit SHA before recording |
| Treating "repo has 5K stars" as evidence the feature works | Stars are a discovery signal, not implementation proof — follow the depth ladder |
Quoting an issue comment without updated_at | Always record both timestamps |
| Assuming anonymous code search will eventually work with retries | It's a hard platform limit (401, no anonymous mode) — check gh auth status first, don't retry into it |
Marking repro.type: actual-run because the README or a maintainer says it works | actual-run means a shown command and its result. A maintainer's assurance is maintainer-confirmed — real evidence, but not a reproduction |
Giving each of the project's own artifacts its own root_id | The project's README, releases, and maintainer comments are one voice: github:owner/repo, however many URLs they occupy |
Collapsing an outside user's issue/comment into github:owner/repo | That's not the project speaking — it's an unaffiliated third party. Use person:<username>, or two independent bug reports get miscounted as one |
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.

