llms.txt
Compare with other sites' llms.txt files →# First Context
> Iheatu Chris Wogu on the values that guide engineering leadership — evidence, contracts, and human-in-the-loop delivery.
Public Ghost content for AI and LLM tooling. Use `/llms-full.txt` for consolidated page and post context.
Append `.md` to any post or page URL to get the content in Markdown (for example, `/example-post.md`).
## Pages
- [About](https://contextfirst.blog/about.md) - Who I am and what First Context is for.
## Posts
- [claude-agent-map: see every agent, and what it’s doing](https://contextfirst.blog/claude-agent-map-see-every-agent.md) - A short note on claude-agent-map: a Claude Desktop pane that shows the main session and every subagent — tier, time, tokens, and tasks — so parallel agent runs stay readable.
- [Inside superlite: the philosophy, the gates, and the decisions that shaped it](https://contextfirst.blog/superlite-philosophy-gates-decisions.md) - How superlite actually works: the coordinator, scripts, hooks, and agent tiers; the plan → lint → dispatch → checkpoint → review loop; and the gates that enforce it.
- [Superlite: hand an agent a spec, come back to tested work](https://contextfirst.blog/superlite-hand-an-agent-a-spec.md) - Why I built superlite, an autonomous build workflow for Claude Code that doesn't ask permission every four minutes: how it works, how to use it, and the benchmarks.
- [Token Compression and the Index Protocol](https://contextfirst.blog/token-compression-index-protocol.md) - Compress the corpus; query the index first
- [Blocker Matrix - Fix What Unblocks the Most Flows](https://contextfirst.blog/blocker-matrix-fix-what-unblocks.md) - Tier by flows unblocked, not ticket age
- [Engineering Delivery Dashboard - Visibility Before Burn-Down](https://contextfirst.blog/engineering-delivery-dashboard.md) - Three layers before you celebrate velocity
- [Outcome-Aligned Teams Around End-to-End Journeys](https://contextfirst.blog/outcome-aligned-teams-journeys.md) - Organize around journeys, not repos
- [Chronic Regression - 3,403 Tickets, 12-Day Re-Emergence Median](https://contextfirst.blog/chronic-regression-3403-tickets.md) - Measure re-emergence, not close rate
- [Graphify - Knowledge Graphs for Monorepo Navigation](https://contextfirst.blog/graphify-monorepo-navigation.md) - Query the graph before you grep
- [Invisible Unicode - Markdown as an Attack Surface](https://contextfirst.blog/invisible-unicode-markdown-attack-surface.md) - Gate smuggling primitives in CI, not in preview
- [Specifications as Execution Contracts](https://contextfirst.blog/specifications-as-execution-contracts.md) - Greenlight before agents implement
- [HTML Dashboards - Metrics Leaders Will Actually Open](https://contextfirst.blog/html-dashboards-leaders-will-open.md) - Static, linkable files beat unread TSV piles
- [393 Failures, 27 Root Causes - Triage by Spread, Not Volume](https://contextfirst.blog/spread-triage-393-to-27.md) - Rank by contexts blocked, not row count
- [Human-in-the-Loop PR Review When AI Writes Faster Than You Read](https://contextfirst.blog/human-in-loop-pr-review.md) - Route review by risk when codegen outruns reading
- [Engineering-Led, Agent-Accelerated](https://contextfirst.blog/engineering-led-agent-accelerated.md) - Humans keep intent; agents execute within bounds
- [Shipping the Agentic Software Delivery Framework](https://contextfirst.blog/shipping-agentic-software-delivery-framework.md) - Methodology as repo, not a slide deck
- [Auditing Analytics Your Agent Wrote](https://contextfirst.blog/auditing-agent-wrote-analytics.md) - Pretty dashboards are not evidence
- [README as Contract - Every Column Defined + Half-Open Windows](https://contextfirst.blog/readme-as-contract-half-open-windows.md) - Two VPs, one CSV, zero shared definition of May
- [The Afternoon We Shipped Three Wrong Metrics](https://contextfirst.blog/afternoon-three-wrong-metrics.md) - Bad numbers erode trust faster than no numbers
- [Parallel Fetch, Raw Cache, and Run Logs](https://contextfirst.blog/parallel-fetch-cache-run-logs.md) - Reproducible pipelines beat live API roulette
- [Your Slack Script Just Shipped](https://contextfirst.blog/script-to-cli-spreadsheet-product.md) - The second team is the product threshold
- [Tiers Beat the Full Matrix](https://contextfirst.blog/e2e-l0-l4-ninety-two-sites.md) - Depth matched to risk, not maximalism
- [Team → Seam → Journey](https://contextfirst.blog/sit-63-workflows-team-seam-journey.md) - 63 workflows so UAT doesn't eat you
- [Prove the Path, Not the POST](https://contextfirst.blog/integration-tests-are-chains.md) - Chains beat endpoints
- [An 857-Line Honesty Machine](https://contextfirst.blog/pr-timeline-script-engineering-metrics.md) - Probe with a script before you productize
- [Green APIs, Broken Handoffs](https://contextfirst.blog/integration-surface-gpr-salesforce.md) - Sample payloads beat architecture slides
- [Fifty Pages, One Map](https://contextfirst.blog/mapping-dtc-launch-mvp-flow.md) - Version the path before you fund the seams
- [Field Notes - Agentic Delivery](https://contextfirst.blog/field-notes-agentic-delivery.md) - Finding where work actually waits
## Optional
- [RSS Feed](https://contextfirst.blog/rss/)
- [Sitemap](https://contextfirst.blog/sitemap.xml)
- [Full content of pages and posts](https://contextfirst.blog/llms-full.txt)