citations
Skill0 starsChanged 55 days agoMITView on GitHub
Standard citation format for human- and LLM-friendly source traceability across all document types. Covers web sources, internal sources (Slack, meetings, voice notes), and chain-of-evidence propagation. This skill defines the format convention — not invoked directly, but referenced by other skills that produce cited documents.
We split this skill into the rules it gives an agent
4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 6 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Use markdown footnotes. Every significant claim gets a footnote.
- Distinctive: line 50
Not all components exist for every source.
- Distinctive: line 71
No link unless a recording/transcript exists.
- Distinctive: line 92
Footnotes travel with claims: When a claim moves from one document to another, its footnote moves too.
“Common” is measured: kept by at least 6 in 10 of the 6 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
Citations
Standard citation format for maintaining source traceability across all document types. Designed to be readable by both humans and LLM agents processing documents downstream.
This skill defines the convention. It is referenced by other skills (pm-spec, research, project-management, etc.) that produce documents with source citations.
All examples below are illustrative. Names, URLs, and quotes are fabricated to demonstrate the format — substitute your own sources.
Footnote Format
Use markdown footnotes. Every significant claim gets a footnote.
Inline reference:
Support agents currently search the knowledge base manually after each call[^call-workflow].Footnote definition (at document end):
[^call-workflow]: Product planning meeting, 2026-01-15. J. Rivera (PM), A. Okonkwo (Eng). "After a call ends, the agent has to go dig through the knowledge base themselves." [View Thread](https://example.slack.com/archives/C00000000/p0000000000000000)Footnote Naming Convention
Short, descriptive slugs: [^org-topic] or [^topic-subtopic]
[^call-workflow]— post-call workflow description[^vendor-sla]— vendor SLA terms[^rivera-scope-v1]— Rivera's v1 scope decision (see the Meetings example below)[^platform-dns]— DNS discussion in #platform
Required Components
Each footnote must include what's available from the source:
- Source context — what kind of source (meeting, Slack thread, web page, voice note, document)
- Date — when it happened or was published
- Participants/author — who said it (when attributable)
- Key claim — what the source says (brief quote or precise paraphrase)
- Link — clickable link when available: URL, Slack thread link, document path
Not all components exist for every source. Include what you have.
Source Types
Web Sources
[^vendor-sla]: Vendor Service Level Agreement page. 99.9% uptime commitment, excluding scheduled maintenance. [View Source](https://www.example.com/legal/sla) (accessed 2026-02-20)Include access date for web sources — content changes.
Slack Threads
[^platform-dns]: #platform thread, 2026-02-10. A. Okonkwo: "Should we make custom DNS a requirement so we don't need to worry about falling back?" [View Thread](https://example.slack.com/archives/C00000000/p0000000000000000)Meetings
[^planning-scope]: Product planning meeting, 2026-01-15. Rivera, Okonkwo, Vance. Rivera: "Let's keep v1 to email only — chat integration is v2."No link unless a recording/transcript exists.
Voice Notes
[^pm-voice-0218]: Voice note, 2026-02-18. J. Rivera. "The key insight is that agents don't want to search — they want articles surfaced for them."Documents
[^prd-v1]: Article Suggestion Engine PRD v1, 2026-01-12. Section: "No-Gos". [View](path/to/prd.md)Issue Tracker Items
[^tracker-update-feb13]: Project update, 2026-02-13. A. Okonkwo. "PR for the per-tenant sending domain is up and approved." [View](https://linear.app/example/project/article-suggestion-engine/updates#project-update-0000)Chain of Evidence
Footnotes travel with claims. When a claim moves from one document to another, its footnote moves too.
Anti-pattern (Broken Chain)
## Decision Document
Support agents search manually after calls (see research-findings.md).Reader can't verify without finding the research document.
Correct Pattern (Preserved Chain)
## Decision Document
Support agents search manually after calls[^call-workflow].
[^call-workflow]: Product planning meeting, 2026-01-15. J. Rivera. "After a call ends, the agent has to go dig through the knowledge base themselves."The footnote is self-contained. Any document, anywhere, can be verified.
Propagation Rules
- Copy footnotes with claims — when moving a claim to a new document, move its footnote too
- Consolidate duplicates — if multiple claims cite the same source, use one footnote
- Never cite intermediate documents as sources — cite the primary source the intermediate document cited
- Verbatim for high-stakes claims — use exact quotes for decisions, constraints, scope changes. Precise paraphrase is fine for context and background.
For Skills That Reference This Convention
Any skill producing cited documents should include in its prompt or instructions:
Citation format: Use markdown footnotes per the citations skill convention.
- Inline: claim text[^slug]
- Definition: [^slug]: Source context, date. Author. "Key quote or paraphrase." [Link](url)
- Footnotes travel with claims across documents (chain of evidence)Or reference this skill directly:
@skills/workflow/citations/SKILL.mdOther versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- ZimoLiao/scholaraio★ 585same name
- matteing/opal★ 59same name
- liangdabiao/social_research_agent★ 45same name
- liangdabiao/simple_claude_deep_research_agent★ 37same name
- blonderoofrat/agent-skills★ 2same name
- polycode-public/claude-code-idea-refinery★ 1same name
- Agentic-Assets/corbis-mcp★ 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 repository43 other files it gives its agents.
- skills/technical/supabase-postgres-best-practices/AGENTS.mdAGENTS.md
- skills/technical/supabase-postgres-best-practices/CLAUDE.mdCLAUDE.md
- direct-handoffSkill
- handoffSkill
- nt-commsSkill
- threadsSkill
- meal-planningSkill
- resume-workflowSkill
- bash-scriptsSkill
- code-reviewSkill
- davinci-resolveSkill
- disk-cleanupSkill
- fastmcp-pythonSkill
- go-cliSkill
- markdown-to-pdfSkill
- pr-createSkill
- qa-planSkill
- rcaSkill
- react-native-best-practicesSkill
- ruby-railsSkill
- skill-creatorSkill
- spec-checkSkill
- supabase-migrationsSkill
- supabase-postgres-best-practicesSkill
- supabase-sqlSkill
- testingSkill
- ui-state-managementSkill
- waydroid-adbSkill
- calcinatioSkill
- exploreSkill
- manual-qaSkill
- nvim-copilotSkill
- orchestrateSkill
- orthogonal-emanationSkill
- product-intent-signal-extractionSkill
- project-management-signal-extractionSkill
- request-researchSkill
- researchSkill
- skill-discoverySkill
- slack-thread-triageSkill
- specSkill
- tmux-sidecarSkill
- youtube-transcribeSkill
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.

