Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/sliday/harn/mcp-validatornpx skills add sliday/harn --skill mcp-validatorgit clone --depth 1 https://github.com/sliday/harnWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00037 | $0.01039 |
| Opus 5 | $0.00018 | $0.00519 |
| Sonnet 5 | $0.00007 | $0.00208 |
| Haiku 4.5 | $0.00004 | $0.00104 |
Grade B, and why
mcp-validator scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
for candidate in "${PROJECT_ROOT}/.claude/settings.local.json" "${PROJECT_ROOT}/.claude/settings.json"; do Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
if ! curl -sf --max-time 2 "$URL" >/dev/null 2>&1; then How it starts
The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MCP Server Health Validator
Generate a bash script at scripts/harness/mcp_health.sh that validates configured MCP servers are reachable. This runs from the SessionStart hook as an informational check (never blocks).
Step 1: Create the health check script
Create scripts/harness/mcp_health.sh and make it executable:
#!/usr/bin/env bash
# SessionStart hook — MCP server health check.
# Informational only — always exits 0.
PROJECT_ROOT="$(cd "$(dirname "$0")/../.." && pwd)"
# Find settings file
SETTINGS=""
for candidate in "${PROJECT_ROOT}/.claude/settings.local.json" "${PROJECT_ROOT}/.claude/settings.json"; do
if [ -f "$candidate" ]; then
SETTINGS="$candidate"
break
fi
done
if [ -z "$SETTINGS" ]; then
# No settings file — nothing to check
exit 0
fi
# Check if jq is available
if ! command -v jq &>/dev/null; then
echo "Harn: jq not found — skipping MCP health check." >&2
exit 0
fi
# Extract MCP server names
SERVERS=$(jq -r '.mcpServers // {} | keys[]' "$SETTINGS" 2>/dev/null)
if [ -z "$SERVERS" ]; then
exit 0
fi
FAILED=()
CHECKED=0
while IFS= read -r name; do
CHECKED=$((CHECKED + 1))
# Check if server has a command
CMD=$(jq -r --arg n "$name" '.mcpServers[$n].command // empty' "$SETTINGS" 2>/dev/null)
if [ -n "$CMD" ]; then
if ! command -v "$CMD" &>/dev/null; then
FAILED+=("$name")
fi
continue
fi
# Check if server has a url
URL=$(jq -r --arg n "$name" '.mcpServers[$n].url // empty' "$SETTINGS" 2>/dev/null)
if [ -n "$URL" ]; then
if ! curl -sf --max-time 2 "$URL" >/dev/null 2>&1; then
FAILED+=("$name")
fi
continue
fi
done <<< "$SERVERS"
if [ ${#FAILED[@]} -eq 0 ]; then
echo "Harn: All MCP servers reachable." >&2
else
for name in "${FAILED[@]}"; do
echo "Harn: MCP server '${name}' unreachable — tools may fail." >&2
done
fi
exit 0
Step 2: Wire into session-start.sh
If scripts/harness/session-start.sh exists, append the MCP health check call. If it does not exist, create it with:
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 149 lines · 37 tokens per session scan B e93b7323e6e6
mcp-validator is a skill published in the GitHub repository sliday/harn (5 stars, last pushed 4mo ago), licensed MIT. It adds 37 tokens to every session and 1,039 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…