Borrowing it
Nothing to install: this file belongs to iliaal/codesage. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/iliaal/codesage/master/.agents/skills/sweep-refs/SKILL.mdgit clone --depth 1 https://github.com/iliaal/codesageWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/iliaal/codesage/sweep-refs)<a href="https://agentmods.dev/skills/iliaal/codesage/sweep-refs"><img src="https://agentmods.dev/badge/skills/iliaal/codesage/sweep-refs/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/iliaal/codesage/sweep-refs"><img src="https://agentmods.dev/badge/skills/iliaal/codesage/sweep-refs.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00043 | $0.02094 |
| Opus 5.5 | $0.00017 | $0.00838 |
| Sonnet 5.5 | $0.00009 | $0.00419 |
| Haiku 4.5 | $0.00004 | $0.00209 |
Grade A, and why
sweep-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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reference sweep
This is the canonical workflow for Codex $sweep-refs and Claude /sweep-refs.
The Claude command only links here. Work from the CodeSage repository root; mirrors
live in ~/ai/codesage_ref/. Accept no argument, YYYY-MM-DD, or
since YYYY-MM-DD; validate the calendar date and reject malformed or future
dates before mutations. Arguments are prompt input, not environment variables.
Deliver recommendations and sweep checkpoints through CodeSage's central br
ledger. Do not create, update, or stage landscape/recommendations Markdown.
Existing notes are read-only bootstrap history. Do not implement recommended
changes, delete legacy notes, commit, push, or post externally without separate
authorization. Preserve unrelated work. Use the writing skill for bead prose;
prefix shell commands with rtk, using rtk proxy for raw/machine-parsed data.
Baseline and migration
- Read the repository's beads protocol and the CLI-contract sections of
~/ai/wiki/tools/beads-review-ledger.md. Runrtk proxy br info --jsonfrom CodeSage and confirm its central store. Never create.beadsin this repo or bypass the wrapper. Capture initial Git status, CodeSage HEAD, and UTC start time. Verify mutation flags against currentbr <command> --help. - Read
rtk proxy br list --status all --deferred --limit 0 --json; checkhas_moreand paginate if needed, including closed/deferred records. Lists return an object withissues;br show ID --jsonreturns a one-element array with comments. Missinglabelsmeans none. Search beyond sweep-labeled records. - Reuse the tracking task labeled
source:sweep-refsandsweep:tracker. Create a P2 taskReference sweep: checkpoint and next reviewonly if none exists. If several exist, reconcile their histories before selecting one. Keep it open: it stores inventory, migration mapping, and the next due date. - Use its latest complete sweep checkpoint as the baseline. Partial runs,
migration-only audits, generic
updated_at, and recommendation edit dates are not completed upstream sweeps. A supplied date overrides the comparison floor, not the carried-item audit. If migration is complete but no full sweep has run yet, use the historical floor preserved in the migration record. Its locally observed mirror HEADs are inventory evidence, not analyzed baselines.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 161 lines · 43 tokens per session scan A 4e7ed74fbe58
sweep-refs is a skill published in the GitHub repository iliaal/codesage (21 stars, last pushed yesterday), licensed MIT. It adds 43 tokens to every session and 2,094 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-09.
Other skills, from other repositories
memtrace-fleet-first
Coordinate fleets of coding agents sharing one repo+branch: publish typed intents, classify edit episodes, and resolve conflicts before they collide. Use FIRST when multiple agents work the same repo+branch, before reading/planning/editing, when joining a fleet or handing work off, and when the user says two agents…
gh-file-issue
Use when filing a new Codewhale GitHub issue: turn a bug or idea into a well-formed, actionable issue with repro, acceptance criteria, labels, and milestone.
gh-treasure-hunt
Hunt the issue/PR queue for highest value-over-risk wins: clean focused community PRs, already-implemented issues to close, safe quick-fixes.
factory-mcp
Use the Warp Factory MCP to hand work to a software factory and collaborate with it — bundle local work and send it to the cloud, find factory tasks from a Slack thread / Linear ticket / description, and pull a task down to test or iterate locally and hand it back.
graph
Whole-codebase structural map of the codanna index as one self-contained HTML disc. Wedges per top-level module, concentric rings with hubs at the centre, hover or click a symbol to light its edge web, search, hide or highlight modules, brush a date range on the ribbon timeline (symbols dated by git blame), heatmap of…
linear
Read and write Linear issues, projects, cycles, comments, and more via Linear's official MCP server. Tools are auto-discovered from the server at runtime.