shared-memory-intake
Curate shared Codex memory when Claude submits findings or Codex reviews, promotes, or maintains task memory.
How do I install this agent skill?
npx skills add https://github.com/ohyeh/agent-scripts --skill shared-memory-intakeIs this agent skill safe to install?
- Gen Agent Trust Hubpass
This skill establishes a shared memory framework for AI agents to store and review task-related information. It includes a Python script for validating the format and security of new entries. The primary security consideration is the potential for indirect prompt injection through the memory ingestion process.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
What does this agent skill do?
Shared Memory Intake
Use ~/.codex/memories/ as the canonical shared knowledge store and
~/.agents/shared-memory-inbox/ as the external submission store. Read
MEMORY.md before opening matching rollout_summaries/ files. Treat dynamic claims
as UNCONFIRMED until verified in the active session.
Actor contract
Require an explicit actor: --as claude or --as codex. Never infer it from the
runtime.
| Actor | Allowed operations | Forbidden operations |
|---|---|---|
claude | read official memory; create one submission in inbox/pending/ | edit or delete official memory, or another submission |
codex | read, review, promote, supersede, maintain | delete historical summaries or silently overwrite a current claim |
Claude: submit
-
Read
MEMORY.md; do not restate an existing current claim as a new submission. -
Create one uniquely named Markdown file under
~/.agents/shared-memory-inbox/pending/with YAML frontmatter:schema_version: 1 source_runtime: claude submitted_at: 2026-07-26T00:00:00Z cwd: /absolute/path source_session: opaque-session-id claim_status: unverified evidence_host: hostname-or-fleet-idevidence_hostis optional for same-host evidence and required when the cited source exists only on another host. -
Keep only reusable conclusions, source references, evidence, and unknowns. Never include credentials, tokens, private keys, cookies, or raw transcripts.
-
Validate before handoff:
python3 ~/.agents/skills/shared-memory-intake/scripts/validate_submission.py /absolute/path/to/submission.md -
A correction is a new submission with
supersedes_submission; never overwrite the earlier proposal.
Codex: review and promote
- Acquire the single-curator lease before promotion:
mkdir ~/.agents/shared-memory-inbox/.curator.lock. If it already exists, stop and report the active-curator conflict. Release it withrmdirafter the outcome is recorded. - Run the validator on the candidate. Reject on schema or secret-pattern failure.
- Inspect the cited source and live-verify every dynamic claim that would enter the
current index. For cross-host evidence, require
evidence_host, verify the live source on that host, and record the curator host plus verification time in the promoted summary. Missing host access remainsUNCONFIRMED. - Re-read
MEMORY.mdimmediately before its update. If its relevant topic changed during review, restart the review instead of merging assumptions. - Promote accepted knowledge to a new immutable
rollout_summaries/*.mdrecord and updateMEMORY.mdas the only current index. Give every changed reusable claim a lifecycle state:active,superseded, orUNCONFIRMED. - Move the candidate to
processed/orrejected/without deleting its audit trail.
Codex: maintain
- Acquire the same single-curator lease and re-read
MEMORY.mdimmediately before every official-memory write; stop on an active-curator conflict. - Scan current index entries for changed paths, versions, models, deployments, or stale references.
- Demote unverified dynamic claims to
UNCONFIRMED; mark replaced claimssupersededwith their replacement reference. - Consolidate duplicate current entries. Do not rewrite historical rollout summaries.
- Re-run the validator on any newly promoted submission and report the paths changed.
Validator
Use scripts/validate_submission.py. It is a narrow preflight gate, not proof that a
claim is true and not a complete secret scanner. A pass authorizes review only; Codex
still owns evidence verification and promotion.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/ohyeh/agent-scripts/shared-memory-intake">View shared-memory-intake on skillZs</a>