agent-observations

A required record of problems, risks, inconsistencies, and improvement suggestions found while working on a code project. It separates open observations from resolved ones.

In plain words
What is it for?
Use it to document blocking issues, recommended follow-up work, and differences between documentation, code, and configuration before committing changes.
Why use it?
It prevents important discoveries from being hidden or forgotten, even when they are outside the current task.

Cursor rule for Cursor

Install

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.

agentmods
npx agentmods add rules/savedpixel/ai-agent-rules-generator/agent-observations
Clone the repo
git clone --depth 1 https://github.com/savedpixel/ai-agent-rules-generator

Made for: Cursor.

Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 713 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00713
Opus 5 $0.00000 $0.00357
Sonnet 5 $0.00000 $0.00143
Haiku 4.5 $0.00000 $0.00071

Measured 2d ago against content hash 84181ea0248d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-observations scanned grade A with 0 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

examples/glitch-payment-gateway/v1.1/cursor/generated/.cursor/rules/agent-observations.mdc · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Observation Logs — Mandatory Disclosure System

Purpose

During any task, you will encounter information the user needs to know. You MUST proactively disclose these findings. Silence is a violation.

Observation logs vs. task logs: Task logs (docs/task/logs/) record what was done (completed work). Observation logs record what was discovered — findings, risks, and anomalies that the user needs to know about, regardless of whether they're related to completed work.


Directory Structure

docs/agent-observations/
├── critical.md
├── recommendations.md
├── anomalies.md
└── closed/
    ├── critical.md
    ├── recommendations.md
    └── anomalies.md
  • Root files = open/in-progress items only.
  • Closed files = append-only resolved archives.

The Three Observation Logs

1. critical.md — Blocking Issues

Data inconsistencies, security misconfigurations, broken assumptions, regression risks, data corruption.

2. recommendations.md — Improvement Suggestions

Follow-up tasks, optimization opportunities, architectural improvements.

3. anomalies.md — Inconsistencies & Drift

Doc/code divergence, unused components, mismatched config values.


Entry Format

| {date} | {source} | {observation} | {impact} | {action} | Open |

When to Write Observations

During Implementation (MANDATORY)

  1. Data consistency — check for order meta integrity, duplicate transaction IDs
  2. Code/doc alignment — verify docs match actual behavior
  3. Assumption validation — confirm WooCommerce API behavior matches expectations
  4. Side effects of changes — check for webhook processing regressions
  5. Things outside your task scope — security concerns, deprecated API usage

During Planning / During Reporting

Log anything unexpected, cross-post findings.


The Observation Gate — BLOCKING

Before you commit:

  1. Review your work.
  2. Classify observations.
  3. Append entries.
  4. If no observations: Observations: none.

Read the full file on GitHub · 111 lines

Changes

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.

  1. 2d ago First seen · 111 lines · 0 tokens per session scan A 84181ea0248d

Subscribe to this mod's changes

agent-observations is a cursor rule published in the GitHub repository savedpixel/ai-agent-rules-generator (7 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 713 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.