Claude Code, Skill
fidelis
Recall locally stored agent memories through the Fidelis MCP server.
Compatibility: Not stated.
Grade B
Minor issues only, no Medium or High. Worth a quick look at the findings below before installing.
Not a guarantee. A hygiene check of the source at the commit shown, not a security audit.
NETWORK_NO_FIXED_HOST SECRETS_PATH_UNCONFIRMED PERMISSION_SHELL_DECLARED SECRETS_ENV_READ OBFUSCATION_FILE_OVER_READ_CAP
Why this grade
- NETWORK_NO_FIXED_HOST Low
Makes a network call whose host is not written as a fixed address.
bench/benchmark.py, line 98
with urllib.request.urlopen(req, timeout=timeout) as resp: - SECRETS_PATH_UNCONFIRMED Low
The Secrets check reports a credential or environment file, but no quoted line shows a recognisable path.
bench/longmemeval_combined_pipeline.py, line 60
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "") - PERMISSION_SHELL_DECLARED Note
Runs shell commands, and the manifest or description says so.
bench/longmemeval_default.py, line 112
"git_head": subprocess.check_output( - SECRETS_ENV_READ Note
Reads an environment variable whose name contains KEY, SECRET, TOKEN, PASSWORD or CREDENTIAL.
bench/longmemeval_combined_pipeline.py, line 60
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "") - OBFUSCATION_FILE_OVER_READ_CAP Note
A file was over the read cap and was not fully read.
bench/analysis-pipeline-2026-04-15.json, line 1
{
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
Note
requests shell access. The manifest names this permission.
PERMISSION_SHELL_DECLARED
bench/longmemeval_default.py, line 112
"git_head": subprocess.check_output(
bench/qa_eval_v3_routing.py, line 378
r = subprocess.run(
bench/run_ablations.py, line 116
proc = subprocess.run(
Show more
bench/run_chunked.py, line 48
result = subprocess.run(cmd, capture_output=False, text=True)
scripts/sustained_write_restore_proof.py, line 117
source_status = subprocess.run(
scripts/sustained_write_restore_proof.py, line 211
"commit": subprocess.run(
scripts/sustained_write_restore_proof.py, line 389
completed = subprocess.run(
src/fidelis/init_cmd.py, line 175
subprocess.run(["launchctl", "unload", str(legacy_plist)], check=False)
Install and update
Pass
No install-time script found, and dependencies are pinned or absent.
Network
Concern
can reach a host that is not written as a fixed address.
NETWORK_NO_FIXED_HOST
bench/benchmark.py, line 98
with urllib.request.urlopen(req, timeout=timeout) as resp:
bench/eval.py, line 92
with urllib.request.urlopen(f"{base_url}{path}", timeout=timeout) as resp:
bench/eval.py, line 109
with urllib.request.urlopen(req, timeout=timeout) as resp:
Show more
bench/longmemeval_combined.py, line 141
with urllib.request.urlopen(req, timeout=120) as resp:
bench/longmemeval_combined_pipeline.py, line 188
with urllib.request.urlopen(req, timeout=120) as resp:
bench/longmemeval_combined_pipeline.py, line 393
with urllib.request.urlopen(req, timeout=30) as resp:
bench/longmemeval_combined_pipeline.py, line 464
with urllib.request.urlopen(req, timeout=15) as resp:
bench/longmemeval_combined_pipeline_dates.py, line 179
with urllib.request.urlopen(req, timeout=120) as resp:
bench/run_all_experiments.sh, line 14
curl -s http://localhost:11434/api/embed -d '{"model":"nomic-embed-text","input":["warm"]}' > /dev/null
bench/run_all_experiments.sh, line 21
curl -s http://localhost:11434/api/embed -d '{"model":"nomic-embed-text","input":["warm"]}' > /dev/null
bench/run_all_experiments.sh, line 28
curl -s http://localhost:11434/api/embed -d '{"model":"nomic-embed-text","input":["warm"]}' > /dev/null
bench/phase-2/run_learned_router.sh, line 5
curl -s http://localhost:11434/api/embed -d '{"model":"nomic-embed-text","input":["warm"]}' > /dev/null
extensions/fidelis-recall.ts, line 74
const response = await fetch(`http://127.0.0.1:${port}/recall_b`, {
src/fidelis/init_cmd.py, line 154
with urllib.request.urlopen(f"http://127.0.0.1:{port}/health", timeout=2) as resp:
Secrets
Concern
reads a credential or environment file.
SECRETS_ENV_READ SECRETS_PATH_UNCONFIRMED
bench/longmemeval_combined_pipeline.py, line 60
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "")
bench/longmemeval_combined_pipeline_dates.py, line 51
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "")
bench/longmemeval_combined_pipeline_flagship.py, line 76
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "")
Show more
bench/longmemeval_combined_pipeline_guard.py, line 75
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "")
bench/longmemeval_combined_pipeline_v31.py, line 96
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "")
bench/longmemeval_combined_pipeline_v32.py, line 102
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "")
bench/longmemeval_combined_pipeline_v33.py, line 98
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "")
bench/longmemeval_combined_pipeline_v33_ablate.py, line 74
DASHSCOPE_API_KEY = os.environ.get("DASHSCOPE_API_KEY", "")
.gitignore, line 10
node_modules/
Agent instructions
Pass
No instruction to ignore earlier instructions or to hide a step.
Obfuscation
Note
a file was over the read cap and was not fully read.
OBFUSCATION_FILE_OVER_READ_CAP
bench/analysis-pipeline-2026-04-15.json, line 1
{
bench/qa_eval_runJ-v33.json, line 1
[
bench/results-default-0.3.0rc1.json, line 1
{
Show more
bench/phase-2/learned_router.pkl, line 1
��� }�(�model��sklearn.ensemble._forest��RandomForestClassifier���)��}�(� estimator��sklearn.tree._classes��DecisionTreeClassifier���)��}�(� criterion��gini��splitter
bench/runs/baseline/per_question.json, line 1
[
bench/runs/baseline-seed2/per_question.json, line 1
[
bench/runs/fix-assistant-text/per_question.json, line 1
[
bench/runs/router-v2/per_question.json, line 1
[
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 9504c4a8f259. requests shell access. The manifest names this permission. can reach a host that is not written as a fixed address. reads a credential or environment file. a file was over the read cap and was not fully read.
Date and source
Date checked:
github.com
Commit graded: 9504c4a8f2592464112ca771d6aaefec07698d23
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.