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 skills add adrielp/ai-engineering-harness --skill observability_driven_developmentgit clone --depth 1 https://github.com/adrielp/ai-engineering-harnessWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/adrielp/ai-engineering-harness/observability_driven_development)<a href="https://agentmods.dev/skills/adrielp/ai-engineering-harness/observability_driven_development"><img src="https://agentmods.dev/badge/skills/adrielp/ai-engineering-harness/observability_driven_development/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/adrielp/ai-engineering-harness/observability_driven_development"><img src="https://agentmods.dev/badge/skills/adrielp/ai-engineering-harness/observability_driven_development.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00089 | $0.01451 |
| Opus 5.5 | $0.00036 | $0.00580 |
| Sonnet 5.5 | $0.00018 | $0.00290 |
| Haiku 4.5 | $0.00009 | $0.00145 |
Grade A, and why
observability_driven_development 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 1mo 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.
This is a copy
92% identical to observability-driven-development — 4 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Observability Driven Development (ODD)
"A feature isn't 'done' when its tests pass; it's done when its trace accurately and legibly narrates the entire request lifecycle." — Observability 4.0 is Inferable, Adriel Perkins
ODD treats the trace as a first-class design artifact. You design the span tree before you write the feature, then run code against a local OTel stack so every change produces a trace you can read in seconds. No guessing, no post-hoc instrumentation.
When to Use ODD
Reach for ODD when any of the following are true:
- The work touches a request lifecycle that crosses async boundaries, services, or external APIs.
- The system has AI agents, MCP servers, or LLM calls (the five domains: context engineering, tool selection & invocation, state management, error recovery, memory access).
- Past production debugging stalled because the existing telemetry didn't narrate what the code was doing.
- You're about to write code where "the trace" is the only honest spec.
If the work is a pure-compute helper, a config tweak, or a CSS change, don't use ODD — it's overhead.
The Inner Loop
Write Code → Instrument (OTel spans) → Run Locally → OTel Collector → Aspire → Observe → [repeat]
Every code change produces a trace. Every trace answers a question. The loop is measured in seconds, not hours. Detail in loop.md.
Workflow
- Write the narrative spec before the implementation. Format: see
narrative.md. Save to
thoughts/shared/telemetry/<feature>.md. Template:thoughts/shared/telemetry/narrative-template.md. - Stand up the local stack if it isn't already running. See local-setup.md. Prefer the Aspire dashboard via Docker.
- Instrument and implement the feature. For SDK setup, span design, and
validation rules, route to
otel_instrumentation. For attribute names, route tootel_semantic_conventions. - Observe in the dashboard after each meaningful change. The trace shape should converge toward the spec.
- Validate with
/validate_telemetry thoughts/shared/telemetry/<feature>.md.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 145 lines · 89 tokens per session scan A e3bbd7f5e345
observability_driven_development is a skill published in the GitHub repository adrielp/ai-engineering-harness (22 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 89 tokens to every session and 1,451 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 92% identical to observability-driven-development, differing in 4 lines, and is treated as a copy.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
deck-retro
Run a fully local agent-deck retrospective over the user's own transcripts, Recall index and logs. Use when a user wants to find recurring failures or take their own finding from investigation through synthetic reproduction, a user-filed issue, a test-first fix and contributor PR handoff. Also use for weekly usage…
incident-response-incident-response
Coordinate production incident triage, mitigation, investigation, recovery, and follow-up. Use when responding to outages or other incidents that require structured operational coordination.
cwv-optimizer
Diagnose and fix Core Web Vitals issues on AEM Edge Delivery Services pages. Goes deeper than generic CWV advice by understanding EDS-specific performance patterns including the 100KB LCP budget, E-L-D loading phases, block rendering behavior, and third-party script impact. Produces specific fixes for LCP, CLS, and…