Skip to content
Back to skills

Nexus Gatekeeper

ASecurity

description: "Formalizes Reality Checker as a hard blocking gate in the task pipeline — a task cannot advance past implementation until the gate is cleared. After implementation, spawns the Reality Checker agent and parses its verdict: PASS sets gate_status=passed (downstream agents spawn), NEEDS_WORK or FAIL sets gate_status=failed and halts the pipeline. Writes gate records to task-store.db. In FAIL scenarios with consecutive failures, escalates to the PD or council. Best for teams that wan...

  • 3 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 29, 2026
ai-agentsgobashsql

Security analysis

A100/100

Scanned September 29, 2026

npx -y skills add Tekkiiiii/the-agency --skill nexus-gatekeeper --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Nexus Gatekeeper?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Nexus Gatekeeper
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tekkiiiii-nexus-gatekeeper/badge)](https://www.skillsdirectory.com/skills/tekkiiiii-nexus-gatekeeper)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
# nexus-gatekeeper

description: "Formalizes Reality Checker as a hard blocking gate in the task pipeline — a task cannot advance past implementation until the gate is cleared. After implementation, spawns the Reality Checker agent and parses its verdict: PASS sets gate_status=passed (downstream agents spawn), NEEDS_WORK or FAIL sets gate_status=failed and halts the pipeline. Writes gate records to task-store.db. In FAIL scenarios with consecutive failures, escalates to the PD or council. Best for teams that want enforcement-grade quality gates without relying on discipline — the gatekeeper makes the workflow self-governing. Also for: tracking recurring quality concerns across tasks and surfacing architectural issues from repeated gate failures."

## Protocol

1. **After implementation:** Spawn Reality Checker agent to assess
2. **Parse verdict:**
   - `PASS` → set `gate_status=passed`
   - `NEEDS_WORK` → set `gate_status=failed`, halt pipeline
   - `FAIL` → set `gate_status=failed`, halt pipeline, escalate
3. **Write gate record** to task store via `task-store` skill
4. **Halt on failed gate:** Do not spawn downstream agents until gate is cleared

## Reality Checker Verdict Reference

- `PASS` — ready to ship, no blocking issues
- `NEEDS_WORK` — significant issues remain, must address before proceeding
- `FAIL` — critical failure, do not continue

## Workflow

```
Implementation Agent
    │
    ▼
Reality Checker Agent
    │
    ├─ PASS → ts-gate passed → spawn downstream agents
    │
    ├─ NEEDS_WORK → ts-gate failed → HALT
    │                    Report to user for remediation
    │
    └─ FAIL → ts-gate failed → HALT
                     Escalate to PD/council
```

## Integration with task-store

The gatekeeper reads from and writes to `~/.claude/task-store.db`:

```bash
# Check gate before spawning
gate=$(sqlite3 ~/.claude/task-store.db \
  "SELECT gate_status FROM tasks WHERE id='$task_id';")
if [ "$gate" = "open" ]; then
  echo "Gate not yet evaluated — run Reality Checker first"
elif [ "$gate" = "failed" ]; then
  echo "BLOCKED — gate failed, see gate_verifier and notes"
else
  echo "Gate passed — safe to advance"
fi
```

## Escalation Path

- Gate `failed` with `NEEDS_WORK`: Report to user with specific issues found
- Gate `failed` with `FAIL`: Escalate to PD or council if multiple consecutive failures
- Gate repeatedly `failed` for the same task: Add to project oversight as recurring quality concern

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…