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 skills add capitalone/context-specs --skill address-feedbackgit clone --depth 1 https://github.com/capitalone/context-specsWrote 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/capitalone/context-specs/address-feedback)<a href="https://agentmods.dev/skills/capitalone/context-specs/address-feedback"><img src="https://agentmods.dev/badge/skills/capitalone/context-specs/address-feedback/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/capitalone/context-specs/address-feedback"><img src="https://agentmods.dev/badge/skills/capitalone/context-specs/address-feedback.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 | $0.00123 | $0.02214 |
| Opus 5.5 | $0.00049 | $0.00886 |
| Sonnet 5.5 | $0.00025 | $0.00443 |
| Haiku 4.5 | $0.00012 | $0.00221 |
Grade A, and why
address-feedback 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 — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
address-feedback
The responder half of the PR review cycle. An automated reviewer posts findings
on the feature PR; you read the unresolved ones, decide what each one is, and act.
You run headless, invoked by the dispatcher as /address-feedback <feature>
inside the feature worktree. One focused pass, then exit.
By the time you run, the feature already works — ./prds/<f>/run-prd-test.sh exits 0
and a PR is open against main. Your job is not to keep building; it is to close out
the reviewer's findings honestly, or to escalate the ones you can't.
The one-way model (read this first — it shapes everything)
The reviewer is non-conversational. Per Anthropic's Code Review docs: "Replying to an inline comment does not prompt Claude to respond or update the PR. To act on a finding, fix the code and push." Two consequences drive this whole skill:
- A Clear fix signals back through the diff, not a thread reply. You fix the cause, commit, push. The reviewer's next push-triggered pass sees the diff and auto-resolves the thread. A courtesy reply does nothing the reviewer can read — so for Clear, do not reply; just push.
- Only a push shrinks the reviewer's unresolved set. A reply (yours or a human's) is invisible to the reviewer and to the dispatcher. So a finding you can only reply to (Ambiguous / Complex / Out-of-Scope) will not converge on its own — it waits for a human, who steers by pushing or merging, not by chatting in threads. That is correct and intended: the dispatcher's round counter marches such a stall to STUCK, which is the right escalation.
You are the agent's self-correction sensor for the bot reviewer. Act only on the automated reviewer's findings. Human review comments belong to the human's own steering loop (merge / STUCK) — leave them alone.
Ground yourself before classifying (you are a coordinator, not a scope authority)
Two sources govern every triage call. Load both before you classify anything:
What ships with it
2 files 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 · 145 lines · 123 tokens per session scan A 2aa829bda849
address-feedback is a skill published in the GitHub repository capitalone/context-specs (44 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 123 tokens to every session and 2,214 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…