MCP, MCP server
vfkb
Decision-grade, git-native memory for AI coding agents.
Compatibility: Not stated.
Grade C
One Medium finding. Read the report before installing.
Not a guarantee. A hygiene check of the source at the commit shown, not a security audit.
SECRETS_PATH_CREDENTIAL
Why this grade
- SECRETS_PATH_CREDENTIAL Medium
Mentions a credential store path (~/.ssh, id_rsa, .npmrc, .aws/credentials, credentials.json, .netrc).
scenarios/l4-purpose.mjs, line 70
// missing ~/.claude/.credentials.json before any scenario runs.
Any High, or two or more Medium findings, give D. One Medium gives C. Low findings or a security Note give B. Each code names a finding behind the letter.
Grades are a hygiene check of the source at the commit shown, on the date checked, not a security audit or a guarantee. The free directory grades are automated. Before release, a PlugVet engineer reviews every paid report and any grade we hold back, including all F grades, against the source code and the evidence we quote.
What was checked
Plain notes from the hygiene checks.
Permissions
Pass
No shell or command execution found, and no extra permission was declared.
Install and update
Pass
No install-time script found, and dependencies are pinned or absent.
Network
Concern
can reach api.typesafe.ai. The description does not name that host. can reach a host that is not written as a fixed address.
NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST
docs/v3/research/probes/jev-respike-2026-10-22/respike.mjs, line 50
const r = await fetch('https://api.typesafe.ai/v1/systemone', {
docs/v3/research/probes/jev-spike-2026-10-22/spike.mjs, line 77
const r = await fetch('https://api.typesafe.ai/v1/systemone', {
scenarios/records/claude-code__claude-haiku-4-5__docker.json, line 196
"output": "Based on the system context, here's the specific guidance:\n\n**Don't rely on the HTTP status code.** The `/healthz` endpoint returns HTTP 200 even when the database is
Show more
scenarios/records/deepseek__deepseek-v4-pro.json, line 237
"output": "The HTTP status code from `/healthz` is **unreliable** - it returns `200` regardless of database health. You must inspect the JSON response body:\n\n**Check:** `response
Secrets
Concern
reads a credential or environment file.
SECRETS_ENV_READ SECRETS_PATH_CREDENTIAL
scenarios/l4-purpose.mjs, line 70
// missing ~/.claude/.credentials.json before any scenario runs.
scenarios/l4-purpose.mjs, line 71
if (HARNESS === 'pi' && PI_MODE === 'docker' && !process.env.DEEPSEEK_TOKEN) {
scenarios/l4-purpose.mjs, line 106
if (!c.claudeAiOauth) throw new Error('no claudeAiOauth in ~/.claude/.credentials.json - log in with host `claude` first');
Show more
scenarios/l4-purpose.mjs, line 108
writeFileSync(join(d, '.credentials.json'), JSON.stringify({ claudeAiOauth: c.claudeAiOauth }), { mode: 0o600 });
scenarios/l4-purpose.mjs, line 298
'-e', `DEEPSEEK_TOKEN=${process.env.DEEPSEEK_TOKEN}`];
scenarios/pi-heal.mjs, line 71
if (!process.env.DEEPSEEK_TOKEN) {
scenarios/pi-heal.mjs, line 93
const tok = process.env.DEEPSEEK_TOKEN;
scenarios/probes/mcp-disconnect.mjs, line 56
const all = JSON.parse(readFileSync(join(homedir(), '.claude', '.credentials.json'), 'utf8'));
scenarios/probes/mcp-disconnect.mjs, line 57
if (!all.claudeAiOauth) throw new Error('no claudeAiOauth block in ~/.claude/.credentials.json');
scenarios/probes/mcp-disconnect.mjs, line 58
writeFileSync(join(cfg, '.credentials.json'), JSON.stringify({ claudeAiOauth: all.claudeAiOauth }));
scripts/tamper-check.mjs, line 280
env: { ...env, TAMPER_TEST_FILES: JSON.stringify(candidates), TAMPER_DECLARED_FILE: declaredFile },
spike/dogfood-smoke.sh, line 18
-v "$HOME/.claude/.credentials.json":/work/.claude/.credentials.json:ro \
Agent instructions
Pass
No instruction to ignore earlier instructions or to hide a step.
Obfuscation
Note
the file cap was reached before every file was read.
OBFUSCATION_FILE_CAP_REACHED
src/curator.ts, line 1
// ACE curator (ADR-0021 / RFC-006) - the maintenance side of auto-distill.
Licence
Pass
A licence file is present (LICENSE).
Maintainer activity
Pass
The latest commit is within the last 365 days.
Note
Hygiene check of commit 57dbb9bff073. can reach api.typesafe.ai. The description does not name that host. can reach a host that is not written as a fixed address. reads a credential or environment file. the file cap was reached before every file was read.
Date and source
Date checked:
github.com
Commit graded: 57dbb9bff07332d9203eb2265790a3a6e532460d
This link goes to the original host. PlugVet does not host or check what you download from there, and installing it is your own decision and risk.