Install verification: runtime verified (L5) · tested on dsh 0.2.1-rc.2 · last checked Install verification details →
What it doesAI
Zero-dependency cross-session memory for AI coding agents that stores lessons on disk, enforces evidence, and auto-injects them into new sessions.
- Zero-dependency memory layer for AI coding agents
- Lessons persisted on disk across sessions
- Evidence-enforced and auto-injected into context
Capability tags are AI-inferred or rule-extracted from the README (see badge) — vocabulary-limited, not install-verified. AI = inferred; README = literal keyword from the readme.
Installation
dsh plugin --profile web add github:GIN0076/cross-session-memory#v0.5.2Install method: GitHub · runtime verified (L5)
Verification & Compatibility
Grouped by plugin version: each card lists the dsh versions actually tested against that artifact version and their install verdicts (L5 wins). Untested combinations are omitted — absence means untested, never assumed compatible. L1–L3 are a plugin-level static archive (independent of the dsh version — the same verdict on every dsh row): they sit in each row's level grid next to L4/L5, marked "static", while their evidence is stored once per plugin inside that row's "checks & evidence" instead of being duplicated per version.
External credentials (plugin-level): none detected
○ The author declares no dsh version requirement — install is not blocked by version.
| plugin \ dsh | dsh 0.2.1-rc.2 current | dsh 0.2.1-alpha.4 | dsh 0.2.1-alpha.2 |
|---|---|---|---|
| v0.5.2 | ✓ | ✓ | ✓ |
v0.5.2github✓ 3 runtime-verified2026-10-19
https://github.com/GIN0076/cross-session-memory
✓dsh 0.2.1-rc.2current2026-10-19L5 · runtime verified
- • Node.js: v24.14.0
- • DSH: 0.2.1-rc.2
- • Plugin artifact: github · https://github.com/GIN0076/cross-session-memory#v0.5.2
ID: cross-session-memory@v0.5.2@dsh0.2.1-rc.2 · profile: l4-sandbox
Issued: 2026-10-19 · expires: 2026-10-26
View checks & evidence
active in loader tree (151 entries)
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
✓dsh 0.2.1-alpha.42026-10-17L5 · runtime verified
- • Node.js: v24.14.0
- • DSH: 0.2.1-alpha.4
- • Plugin artifact: github · https://github.com/GIN0076/cross-session-memory#v0.5.2
ID: cross-session-memory@v0.5.2@dsh0.2.1-alpha.4 · profile: l4-sandbox
Issued: 2026-10-17 · expires: 2026-10-24
View checks & evidence
active in loader tree (151 entries)
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
✓dsh 0.2.1-alpha.22026-10-15L5 · runtime verified
- • Node.js: v24.14.0
- • DSH: 0.2.1-alpha.2
- • Plugin artifact: github · https://github.com/GIN0076/cross-session-memory#v0.5.2
ID: cross-session-memory@v0.5.2@dsh0.2.1-alpha.2 · profile: l3-fullscan
Issued: 2026-10-15 · expires: 2026-10-22
View checks & evidence
active in loader tree (151 entries)
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
Only the latest 3 published versions are shown; source-snapshot (commit hash) groups, unstated groups and older versions are omitted — the full archive stays in the data layer (snapshot/unstated groups appear only when a plugin has no published version).
Author badges
Embed the official badges in your README to showcase security & install-test status:
`[](https://www.dsh.so/artifact/cross-session-memory/)``[](https://www.dsh.so/artifact/cross-session-memory/)`Security Report
Two automated layers, neither a manual review: a static scan (per-plugin rule checks) and a model-driven dynamic security audit that covers only the subset static-flagged at warn or above and never executes plugin code.
Static low, dynamic medium — stricter wins → overall medium.
v0.5.2dsh manifestv0.5.2git2026-10-21v0.5.2same versionclean · gitVet is an AST static scan (npm artifact or git source).
show static findings (4)
- warningchildProcess · code-exec
install/smoke.mjs:37const r = spawnSync(process.execPath, ['--check', join(copy, f)], { stdio: 'inherit' }) - warningchildProcess · code-exec
install/smoke.mjs:47const setup = spawnSync(process.execPath, [join(copy, 'install', 'setup.mjs'), '--with-sample'], { cwd: copy, encoding: 'utf8' }) - warningspawnShell · shell
plugin/dsh-memory/tools/mem.mjs:2075const r = spawnSync(cmd, { shell: true, encoding: 'utf8', timeout: 5000, cwd: memoryRoot() }) - warningspawnShell · shell
tools/mem.mjs:2075const r = spawnSync(cmd, { shell: true, encoding: 'utf8', timeout: 5000, cwd: memoryRoot() })
show 2 audit findings
- mediumglobal-sync deletes shared ~/.memory-global mirrors it does not own (missing sourceWorkspace => deleted)
tools/mem.mjsThe whole point of ~/.memory-global is cross-workspace sharing, but this engine never writes sourceWorkspace into mirrored entries. So every mirror created by any workspace has owner === undefined, the `if (owner && ...)` guard falls through, and unlinkSync removes it. Consequently the last workspace to run `mem global-sync` silently deletes every other project's global memory mirrors. The root is overridable via DSH_MEMORY_GLOBAL_DIR, so the deletion target directory is environment-controlled.
- lowverifyEntry runs entry-supplied command through spawnSync with shell:true
tools/mem.mjsThe command string comes from an entry's frontmatter (metadata.verification.command), i.e. memory content, and is handed to a real shell. In practice it is not exploitable: both regexes are fully anchored, the path class excludes all shell metacharacters, and the line-based frontmatter parser cannot emit embedded newlines into the value, so only `node --check <path>.mjs` / `git status --porcelain` can ever reach sh. The shell flag is still an unnecessary attack surface (e.g. a future parser change that admits whitespace/newlines would become command injection).
Local-only memory engine (no network, no exfil); two real issues: global-sync deletes other workspaces' shared mirrors, and a whitelisted shell:true exec path.
Rule: the dynamic audit can only raise risk, never lower it (raise-only, never an exemption); a wrong raise is withdrawn by human review and the tier falls back to the static verdict. Overall verdict = the stricter of the two layers.
Disclaimer: the static scan and the dynamic security audit are both automated analysis, not a security guarantee. Always review what you install.
Activity
Last commit 2026-10-01 · activity: Active
• Repo created: 2026-09-22
• Stars: ★ 0 · Forks: 0
• Health: Active — committed within last 30 days
Source
Author README
View on GitHub ↗The author's original README from the repository — fetched live, unedited (the English page loads the default README).
