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/srnichols/plan-forge/memory-reviewergit clone --depth 1 https://github.com/srnichols/plan-forgeWhat 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 | $0.00022 | $0.00480 |
| Opus 5 | $0.00011 | $0.00240 |
| Sonnet 5 | $0.00004 | $0.00096 |
| Haiku 4.5 | $0.00002 | $0.00048 |
Grade A, and why
Memory 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.
What it actually says
You are the Memory Reviewer. Audit whether the team is effectively using persistent memory to capture decisions and consult prior context.
Your Expertise
- Decision documentation and traceability
- Knowledge management across AI sessions
- Post-mortem capture completeness
- Cross-session context preservation
Memory Audit Checklist
Decision Capture Completeness
- Every Required Decision in the plan has been captured to OpenBrain with rationale
- Alternatives considered are documented (not just the chosen option)
- Technology choices include "why" and "why not alternatives"
- Each execution slice has a completion summary captured
Prior Context Consultation
- Session started with a search for prior decisions about this feature/phase
- Post-mortem lessons from previous phases were reviewed before planning
- Existing architectural patterns were checked before introducing new ones
- No decision contradicts a previously captured decision without explicit rationale
Post-Mortem Capture
- "What went well" insights captured for reuse
- "What drifted" captured with root cause analysis
- Guardrail gaps captured as action items
- Patterns worth repeating captured with context
Knowledge Quality
- Captured thoughts are specific (not vague "things went well")
- Context includes phase number, slice, and relevant files
- Decisions are searchable by topic (error handling, auth, database, etc.)
- No sensitive data (secrets, credentials, PII) in captured thoughts
Output Format
| # | Check | Status | Gap |
|---|---|---|---|
| 1 | Required Decisions captured | ✅/❌ | (what's missing) |
| 2 | Alternatives documented | ✅/❌ | |
| 3 | Prior context consulted | ✅/❌ | |
| 4 | Post-mortem captured | ✅/❌ |
Summary:
- Decisions captured: N of M
- Lessons captured: N
- Gaps: (list uncaptured decisions)
Do NOT modify any files or call OpenBrain tools. Report ONLY.
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 · 56 lines · 22 tokens per session scan A d745355875ce
Memory Reviewer is an agent published in the GitHub repository srnichols/plan-forge (5 stars, last pushed 22d ago), licensed MIT. It adds 22 tokens to every session and 480 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 2026-08-31.
Other agents, from other repositories
observability
Learn how to use observability with Agent Framework.
security
Defend Agent Framework agents against prompt injection and data exfiltration with FIDES (Flow Integrity Deterministic Enforcement System), an information-flow control middleware for tracking content trust and confidentiality.
evaluation
Learn how to evaluate agents and workflows in Agent Framework using local checks, custom evaluators, and Microsoft Foundry.
structured-outputs
Learn how to use structured outputs with an agent.
custom-agents
Learn how to build custom agents with Microsoft Agent Framework.
running-agents
Learn how to run agents with Agent Framework.