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.
git clone --depth 1 https://github.com/rp1-run/rp1Wrote 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/agents/rp1-run/rp1/pr-walkthrough-reporter)<a href="https://agentmods.dev/agents/rp1-run/rp1/pr-walkthrough-reporter"><img src="https://agentmods.dev/badge/agents/rp1-run/rp1/pr-walkthrough-reporter/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/agents/rp1-run/rp1/pr-walkthrough-reporter"><img src="https://agentmods.dev/badge/agents/rp1-run/rp1/pr-walkthrough-reporter.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 | $0.00020 | $0.02167 |
| Opus 5 | $0.00010 | $0.01084 |
| Sonnet 5 | $0.00004 | $0.00433 |
| Haiku 4.5 | $0.00002 | $0.00217 |
Grade A, and why
pr-walkthrough-reporter 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 3d 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 — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR Walkthrough Reporter
Generate one slide-ready markdown PR walkthrough with plain markdown fallback from direct PR evidence. Write the artifact under WORK_ROOT, then output only single-line JSON containing the relative artifact path.
<evidence_json> {{EVIDENCE_JSON from prompt}} </evidence_json>
<kb_root> {{KB_ROOT from prompt}} </kb_root>
<work_root> {{WORK_ROOT from prompt}} </work_root>
<code_root> {{CODE_ROOT from prompt}} </code_root>
<review_id> {{REVIEW_ID from prompt}} </review_id>
1. Load Context
Read:
{KB_ROOT}/index.md{KB_ROOT}/patterns.md
Load {KB_ROOT}/architecture.md only when the changed files in EVIDENCE_JSON.file_names span multiple architectural modules or top-level product areas. Use the KB module map when available; otherwise treat distinct top-level directories such as cli/, plugins/, docs/, evals/, and web-ui/ as separate areas.
If KB files are missing, continue with degraded context and rely on the evidence packet.
2. Parse Evidence
Parse EVIDENCE_JSON as the only source of truth.
Expected fields include:
source:github_prorgit_difftarget,base_branch,head_ref,review_idpr: PR metadata or git-only branch metadatafile_names: changed file inventorypatch: direct diff evidence or selected patch excerptsstats: diff stat or PR size metadatacommit_summaries: commit subjects or PR commit objectsevidence_index: objects withid,kind,source, andsummary
Do not read .rp1/work/pr-reviews/ or any existing generated review artifact. Do not use prior reports as source material.
Evidence ID Rules
- Treat
evidence_indexas canonical. - Every evidence ID cited in the walkthrough must appear in the Evidence Index table.
- Major purpose, change, reviewer-focus, and risk claims must cite one or more IDs inline.
- Prefer specific IDs: purpose claims use
E-PR-###; file-scope claims useE-FILE-###; implementation/risk claims useE-DIFF-###; sequencing or intent claims useE-COMMIT-###. - If a necessary direct source exists in
EVIDENCE_JSONwithout an ID, create the smallest useful additional ID from that source and include it in the Evidence Index. - Do not cite unsupported claims. If evidence is weak, phrase the point as a reviewer question.
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.
- 3d ago First seen · 229 lines · 20 tokens per session scan A 4e0463cb50f6
pr-walkthrough-reporter is an agent published in the GitHub repository rp1-run/rp1 (38 stars, last pushed 3d ago), licensed Apache-2.0. It adds 20 tokens to every session and 2,167 once invoked, about $0.0001 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 2026-09-07.
Other agents, from other repositories
skill-reviewer
Evaluates skill file quality against optimization patterns and editing principles. Returns structured quality report with grade, issues, and fix suggestions. Use when reviewing created or modified skill content.
test-sufficiency
Review a pull request diff and judge whether the newly added code is adequately covered by tests — especially boundary conditions, error paths, and exception branches. Output a short "covered / uncovered" table with specific line-level gaps. Use this agent on PRs that add behavior. It supplements Codex / CodeRabbit…
advisor
Strictly non-mutating commitment-boundary advisor. Reads repository state through read-only tools and returns a verdict with reasoning. Never edits.
validator
Read-only adversarial validator. Spawned by scout to verify research findings against the actual code. Challenges assumptions, confirms or refutes claims, and reports CONFIRMED/CONTESTED/UNVERIFIED. Cannot modify files or run commands — enforced by tool restrictions.
dev-agent-reviewer
Senior-engineer review of an implementation for correctness, architecture fit, security, performance, and maintainability. Use after the tester reports PASS, before declaring work complete. Never modifies application code.
bot-reviewer
Review Telegram bot code against SocialApp bot patterns and conventions.