Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/mdsakalu/agent-plugins/refsgit clone --depth 1 https://github.com/mdsakalu/agent-pluginsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00008 | $0.00517 |
| Opus 5 | $0.00004 | $0.00259 |
| Sonnet 5 | $0.00002 | $0.00103 |
| Haiku 4.5 | $0.00001 | $0.00052 |
Grade A, and why
refs scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Search and display references across all workspaces (PRs, Jira tickets).
Resolve the metarepo path by reading ~/.claude/workspace-config.json and expanding ~ to $HOME.
The user's request: $ARGUMENTS
Steps
- Read
~/.claude/workspace-config.jsonto get the metarepo path. Expand~to$HOME. - Scan all workspace metadata files:
<metarepo>/notes/**/workspace.json- This covers all repos and all workspaces (active + retired)
- For each workspace.json, extract the
referencesarray. - Build a combined list with workspace context: repo name, workspace name, active/retired status.
- If the user requested filtering:
- By type:
--type pror--type jira - By repo:
--repo my-repo - By status: active-only or retired-only (check
retiredfield in workspace.json) - By search term: filter refs matching a keyword
- By type:
- Optionally enrich PR references with live status:
- For each repo, read
.repo.jsonfor GitHub owner/repo - Run:
gh pr view <number> --json state,title --repo <owner>/<repo>(limit to 10 PRs) - Show state: OPEN, CLOSED, MERGED
- For each repo, read
- Format output grouped by workspace:
my-repo/epic-auth-refactor (active) - PR #42: Add user authentication [MERGED] - Jira PROJ-100 my-repo/pr-99 (retired) - PR #99: Fix pagination bug [CLOSED] - Show summary: total references, breakdown by type, breakdown by repo.
Important
- Always resolve metarepo from
~/.claude/workspace-config.json, never hardcode paths. - References are in
notes/<repo>/<workspace>/workspace.json(source of truth). - Scan all repos and workspaces, not just the current one.
- Limit
gh pr viewcalls to avoid rate limiting (max 10 enrichment calls). - Handle cases where workspace.json exists but has no references.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 46 lines · 8 tokens per session scan A 319335f1a906
refs is a command published in the GitHub repository mdsakalu/agent-plugins (4 stars, last pushed 2mo ago), licensed MIT. It adds 8 tokens to every session and 517 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.