Install verification: Ecosystem plugin: installs via dsh plugin add but declares no dsh.bundle and mounts no plugin capability — the L1-L5 install ladder does not rate it; the security scan still runs.Install verification details →
What it doesAI
A DSH agent preset for problem-driven, Lab-centered project learning with handouts, auto-graded stages, reviews, and write-permission guardrails.
- One Lab = one stage: handout, rules, runnable autograder, and retrospective
- Same methodology shipped for Pi, OpenCode, and DSH harnesses
- Persona plus model-level meta-rules block out-of-scope writes and protect grade files
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:mingger77/project-learning-helperInstall method: GitHub · runtime test inconclusive (L5 unknown)
Verification & Compatibility○ Ecosystem plugin
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
package manifest not collected — the declared range is unknown.
| plugin \ dsh | dsh 0.2.1-rc.2 current | dsh 0.2.1-alpha.4 | dsh 0.2.1-alpha.2 |
|---|---|---|---|
| da623b | ○ | ○ | ○ |
| c0a106 | · | · | · |
da623bgithub○ 3 ecosystem plugin2026-10-19
https://github.com/mingger77/project-learning-helper
○dsh 0.2.1-rc.2current2026-10-19Ecosystem plugin
- • Node.js: v24.14.0
- • DSH: 0.2.1-rc.2
- • Plugin artifact: github · https://github.com/mingger77/project-learning-helper (version unstated)
ID: project-learning-helper@da623b@dsh0.2.1-rc.2 · profile: l4-sandbox
Issued: 2026-10-19 · expires: 2026-10-26
View checks & evidence
not present in web-profile loader inventory
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
○dsh 0.2.1-alpha.42026-10-18Ecosystem plugin
- • Node.js: v24.14.0
- • DSH: 0.2.1-alpha.4
- • Plugin artifact: github · https://github.com/mingger77/project-learning-helper (version unstated)
ID: project-learning-helper@da623b@dsh0.2.1-alpha.4 · profile: l3-fullscan
Issued: 2026-10-18 · expires: 2026-10-25
View checks & evidence
not present in web-profile loader inventory
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
○dsh 0.2.1-alpha.22026-10-14Ecosystem plugin
- • Node.js: v24.14.0
- • DSH: 0.2.1-alpha.2
- • Plugin artifact: github · https://github.com/mingger77/project-learning-helper (version unstated)
ID: project-learning-helper@da623b@dsh0.2.1-alpha.2 · profile: l4-sandbox
Issued: 2026-10-14 · expires: 2026-10-21
View checks & evidence
not present in web-profile loader inventory
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
c0a106local2026-10-14
Ecosystem plugin: installs via dsh plugin add but declares no dsh.bundle and mounts no plugin capability — its level reads L5 · Ecosystem plugin.
Author badges
Embed the official badges in your README to showcase security & install-test status:
`[](https://www.dsh.so/artifact/project-learning-helper/)`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.
—d8d2626035c5git2026-10-21d8d2626035c5same versionclean · gitVet is an AST static scan (npm artifact or git source).
show static findings (9)
- warningshellExec · shell
opencode-variant/.opencode/skills/project-lab-method/references/check-rules.mjs:166const out = execFileSync("git", ["diff", "--numstat", "HEAD"], { - warningchildProcess · code-exec
opencode-variant/.opencode/skills/project-lab-method/references/grade.mjs:63const result = spawnSync(test.command, { - warningshellExec · shell
pi-plugin/.pi/skills/project-lab-method/references/check-rules.mjs:166const out = execFileSync("git", ["diff", "--numstat", "HEAD"], { - warningchildProcess · code-exec
pi-plugin/.pi/skills/project-lab-method/references/grade.mjs:63const result = spawnSync(test.command, { - warningshellExec · shell
project-learning-preset/skills/project-lab-method/references/check-rules.mjs:166const out = execFileSync("git", ["diff", "--numstat", "HEAD"], { - warningchildProcess · code-exec
project-learning-preset/skills/project-lab-method/references/grade.mjs:63const result = spawnSync(test.command, { - warningmanifest-fs-unfenced · sandbox
project-learning-preset/agent.cordis.ymlMutating filesystem tool has no explicit fs backend in realm - warningmanifest-search-unfenced · sandbox
project-learning-preset/agent.cordis.ymlSearch tool reads without an explicit fs mount - warningmanifest-shell-unfenced · sandbox
project-learning-preset/agent.cordis.ymlShell tool without a sandbox runner or policy in realm
show 2 audit findings
- mediumAutograder shell-executes model-authored commands from lab.config.json without allowlist or confirmation
pi-plugin/.pi/skills/project-lab-method/references/grade.mjs:63tests[].command strings are produced by the agent via learning_lab and persisted in lab.config.json (later frozen by the guard). learning_grade then runs `node autograder/grade.mjs`, which pipes every command through /bin/sh with the user's inherited environment (env: { ...process.env }). Unlike `git commit`, this execution path has no user confirmation, so any content that steers the model (a hostile README, an injected source comment, a poisoned issue) can land as an arbitrary shell command that the tool then runs on the user's behalf. Same copies exist at opencode-variant/.opencode/skills/p
- lowinstall.sh recursively deletes <argv1>/.pi with no target validation or backup
pi-plugin/scripts/install.sh:15dst is built directly from the first positional argument (dst="$target/.pi") with no normalization or sanity check. A mistyped or shell-expanded target (for example an accidental $HOME or a path with a stray variable) causes an unrecoverable recursive delete of that directory's .pi content once --force is supplied, before any copy is attempted.
Learning-mode plugin whose write guard is sound; the real risk is the autograder shell-executing model-authored lab.config.json commands, plus an unvalidated rm -rf in install.sh.
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 2026-09-18 · activity: Stable
• Repo created: 2026-09-06
• Stars: ★ 1 · Forks: 0
• Health: Stable — committed within 90 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).
