Borrowing it
Nothing to install: this file belongs to stellar-experimental/stellar-raven. 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/stellar-experimental/stellar-raven/main/.agents/skills/improvements-pipeline/SKILL.mdgit clone --depth 1 https://github.com/stellar-experimental/stellar-ravenWrote 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/stellar-experimental/stellar-raven/improvements-pipeline)<a href="https://agentmods.dev/skills/stellar-experimental/stellar-raven/improvements-pipeline"><img src="https://agentmods.dev/badge/skills/stellar-experimental/stellar-raven/improvements-pipeline/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/stellar-experimental/stellar-raven/improvements-pipeline"><img src="https://agentmods.dev/badge/skills/stellar-experimental/stellar-raven/improvements-pipeline.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00077 | $0.04709 |
| Opus 5 | $0.00039 | $0.02354 |
| Sonnet 5 | $0.00015 | $0.00942 |
| Haiku 4.5 | $0.00008 | $0.00471 |
Grade A, and why
improvements-pipeline 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 10d 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 — 338 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Improvements pipeline
Use this skill when work touches improvements/: filing upstream findings, changing statuses,
running probes, updating intake targets, regenerating the index, or interpreting lint failures.
Read improvements/README.md first; it is the charter.
Status lifecycle
Statuses are evidence bars:
proposed: plausible finding, usually from an eval transcript or review, but not yet live-proven.verified: live re-execution proves the finding against the upstream surface; record the command, service response, result stamp, or other reproducible evidence.reported-upstream: filed with the service owner; evidence must include the durable filed ref.declined-upstream: the trigger still reproduces but the owner explicitly declined, classified it as legacy, or rejected the recommendation as overfit. Keep the record, include the decline ref and a non-emptydisposition, do not re-file/pester, and revisit only on drift or materially new evidence.fixed-upstream: live re-check confirms the upstream fix. If a residual remains, create or update a successor finding instead of stretching the old one. This is a transient deletion-candidate queue, not an archive state.
The terminal resolved transition is deletion from the active collection plus an entry in
improvements/resolved.json. IDs are unique across the active tree and resolved ledger and are never
reused; choose the next numeric id from the maximum across both. Git history, the resolved receipt,
and the upstream resolution comment preserve evidence without retaining stale finding files.
Findings are for upstream service/data/content/spec gaps only. Own-repo fixes go to .agents/TODO.md.
GitHub state is not truth by itself. An upstream issue being closed or a PR being merged is
evidence to inspect; a finding moves to fixed-upstream only after re-running the original
trigger, recurrence probe, or live repro and observing the fix.
Filing workflow
Dedupe first: search existing finding ids, titles, and recurrences before adding a file. A new finding needs a concrete owner-facing recommendation, quantified prevalence when possible, and evidence a stranger can reproduce. Use the next id in the service prefix sequence:
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.
- 10d ago First seen · 338 lines · 77 tokens per session scan A 356a25be1491
improvements-pipeline is a skill published in the GitHub repository stellar-experimental/stellar-raven (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 77 tokens to every session and 4,709 once invoked, about $0.0004 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 skills, from other repositories
increment
Plan a unit of work as a SpecWeave increment - spec.md with Problem, Scope, numbered ACs and an Approach, plus tasks.md. Use when starting a feature, bug, hotfix or refactor.
sw-task
Claim, finish, skip and release SpecWeave tasks via the append-only ledger so many agents share one increment. Use for "claim a task", "mark task done", "what next", "who owns T-03".
sw-do
Implement a SpecWeave increment task by task through the ledger, with evidence per task and a verified close. Use for "implement this", "start working", "continue the increment", "keep going".
sw-increment
Open a SpecWeave increment - a numbered folder with spec.md (problem, scope, numbered ACs, approach) and tasks.md - before any code. Use for "plan a feature", "new increment", "let's build X".
do
Work an increment task by task through the ledger: task next, claim, implement, commit, task done with evidence. Use when saying "implement", "start working", or "continue increment".
done
Close an increment: ledger check, specweave verify, optional review, then specweave complete. Use when all tasks are done and saying "close increment", "we are done", or "finish up".