poordjaevinCAUTION
Djævin: calibrated local-first decision layer with a Devin ACP backend - no API key needed
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Part of the [DEVIN ecosystem](https://github.com/Icaro0310/awesome-devin) Track: Control · Nature: product For: AI engineers, QA engineers Interface: CLI
10e347b5dba9OBSERVED · 2026-10-11Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add poordjaevin -- None poordjaevin==0.1.1
Exposed tools (10)
10 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
classify | read | Pick exactly one of `options` for `text` (optional `context`). |
decide | read | Answer a batch of typed questions in one call. Each item is |
djaevin_calibrate | read | Report: mean confidence on disagreed decisions per tool and a |
djaevin_usage | read | Model/quota monitor: decisions and measured ACP turn cost, grouped |
gate | read | Advisory opinion: risk that `action` moves money or destroys data |
judge | read | Yes/no opinion on `statement` (optional `context`). Returns |
keepalive | read | Renew the model |
mark_disagreement | read | Mark a logged decision (its `decision_id`) as wrong. With |
rate | read | Rate `text` on `scale`: a short numeric range ( |
usage | read | Return local inference usage counters for this server process. |
Trust audit
CAUTIONgrade B · trust 89/100 Install with care. The audit found things worth knowing before you trust its output.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (6 observation(s))
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (2)
".env" ... urllib.request
.pre-commit-hooks.yaml
Gates applied: no_behavioural_pass.
10e347b5dba9full audit observations/trust-audit/mcp-server/icaro0310__poordjaevin.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-11 | 10e347b5dba9 | CAUTION | B | 89 | first audit |
Questions
What is the poordjaevin MCP server?
Djævin: calibrated local-first decision layer with a Devin ACP backend - no API key needed
What tools does poordjaevin expose?
10 in total: 10 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is poordjaevin safe to connect to an agent?
With care. The audit graded it B (89/100) and found 2 things worth knowing before you trust this server, listed below with the exact line each was found on.
What credentials does poordjaevin need?
It reads DEVIN_CREDENTIALS_PATH and TYPESAFE_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does poordjaevin run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as poordjaevin.
How current is this page?
The grade is for one exact copy of the source (10e347b5dba9), read on 2026-10-11. The repository is watched and re-audited when it changes.