agent-memory-projectSAFE
Coding/repo add-on overlay for the agent-memory framework — wizards, doc-gen, QA, fleet, localization, push/pull. Depends on the memory core (agent-memory-system); composes on top for coding agents.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
The coding/repo add-on overlay for the agent-memory framework.
This repo holds the coding- and repository-oriented procedures that sit on top of the memory core: the coding shell (awaken/wrap-up), project memory, and push/pull. The wizards + QA moved to `agent-memory-wizards` (2026-10-09), the fleet to `agent-memory-fleet` (2026-10-08), and the localization lane to `agent-memory-local` (2026-10-09) — on that last day this repo was renamed agent-memory-project and made central-only. A coding agent installs the memory core plus this overlay; a plain chat agent installs the core alone.
Relationship to the memory core
- Standalone, independent repo — a peer of the core (`agent-memory-system` /
control-files). It is not a git submodule of the core or of the parent aggregator. - One-way dependency: this overlay references the core; the core never references this overlay by name (enforced by the core's
check-core-invariant.shguard). - Additive composition, not override: the consuming agent is the composition point.
procedures/awaken-coder.mdsimply orchestrates "run the core `/awaken-agent`, then coding reasoning + localized-home + orientation map + fleet." Nothing here overrides a core procedure.
Contents (procedures/)
- Awakening overlay:
awaken-coder.md(composes the core awaken, then hands off toagent-memory-localwhen a project is localized) - Lifecycle:
project-wrap-up.md(composes the core/wrap-up, then push + map-orientation) - Wizards (moved out 2026-10-09): the wizard protocols +
implement-plannow live in the sibling repo `agent-memory-wizards` (hermod-wizards
c63d155a87c4OBSERVED · 2026-10-13Host compatibility
What the documentation claims. We have not run a compatibility test.
| Host | Status | Notes |
|---|---|---|
| claude-code | mentioned | |
| claude-desktop | mentioned | |
| codex | mentioned | |
| cursor | mentioned |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (4 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (3)
- **Implementation**: start the process; connect with `claude mcp add --transport http hermod http://127.0.0.1:<port>/mcp` or a `.mcp.json` entry; exercise `ping`, `list_procedures`, `read_procedure("- **Testing Log**: `tests/test_app.py` (4 tests) asserts `/health` returns exactly the service identity, that a real `initialize` POST to `/mcp` comes back 200 with the protocol version and the server
Gates applied: no_behavioural_pass.
c63d155a87c4full audit observations/trust-audit/skill/alvseek__agent-memory-project.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-13 | c63d155a87c4 | SAFE | B | 89 | first audit |
Questions
What does the agent-memory-project skill do?
Coding/repo add-on overlay for the agent-memory framework — wizards, doc-gen, QA, fleet, localization, push/pull. Depends on the memory core (agent-memory-system); composes on top for coding agents.
Is agent-memory-project safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can agent-memory-project access on my machine?
The audit observed that it reads or writes files. Each of those is consistent with what it says it does. Secrets in the source: none found.
Which assistants does agent-memory-project work with?
Its documentation mentions claude-code, claude-desktop, codex and cursor. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (c63d155a87c4), read on 2026-10-13. The repository is watched, and a new audit runs when it changes — this is the first audit.