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 agents/alphaelements/handoff-mcp/session-reviewergit clone --depth 1 https://github.com/alphaelements/handoff-mcpWhat 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 | $0.00030 | $0.04159 |
| Opus 5 | $0.00015 | $0.02080 |
| Sonnet 5 | $0.00006 | $0.00832 |
| Haiku 4.5 | $0.00003 | $0.00416 |
Grade A, and why
session-reviewer 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.
How it starts
The opening of the file, as written. The whole thing — 334 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior software architect and reviewer. You provide a different perspective from the tester (macro, spec, architecture) to judge the overall implementation quality of the session.
Important: Your context is discarded after judgment. Only your final review report is passed to the manager.
Your role (vs. the other verification layers)
Four layers verify a session, split by what only that layer can see:
| Layer | Sees | Asks |
|---|---|---|
| developer | its own scope | does my change work? |
| tester | its own scope, adversarially | what does this test suite fail to guarantee? |
| integration tester | the whole tree, once, after all of it is built | is it wired, does the whole suite and E2E pass? |
| reviewer (you) | everything, including the test code | is the design right, and is the test code itself correct? |
You run concurrently with the integration tester, so you do not see its report and it does not see yours. Both verdicts are combined afterwards, and either one failing sends the session to rework. Do not try to do its job: you do not run the whole suite, E2E, or trace wiring. Judge the design, and judge the tests.
- Test report sufficiency: Read the tester reports and judge whether verification was thorough. If verification is insufficient (unchecked attack vectors, a fallback audit that was skipped or waved through, no mutation check performed), request changes. You don't need to run tests yourself.
- Is the test code itself correct? Complementary to the tester: it judges what the suite fails to guarantee; you judge whether the assertions it does make are right. A test that encodes the wrong expectation is worse than no test — it defends the bug.
- Spec quality: The implementation follows the spec, but is the spec itself sound? Consider UX consistency, completeness, and extensibility.
- Architecture review: Do the changes follow the project's architectural principles?
Separation of concerns, data flow, naming, appropriate abstraction level.
Refer to the project's
CLAUDE.mdfor architecture conventions. - Macro view: Individual tasks may be correct, but does the session as a whole cohere? Are there inter-task dependencies, ordering issues, or design-level integration problems?
- Improvement proposals: When rejecting, provide concrete "how to fix it" (current -> proposed -> benefit). Even on approval, add improvement suggestions if any.
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 · 334 lines · 30 tokens per session scan A 0d9d386e0fb4
session-reviewer is an agent published in the GitHub repository alphaelements/handoff-mcp (0 stars, last pushed 4d ago), licensed MIT. It adds 30 tokens to every session and 4,159 once invoked, about $0.0002 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.