Examples - Programmer.ie: Modern AI programming

Loading examples…

Search requires JavaScript. Browse examples below.

Watch the agent loop call a tool — Pi Agents

Resolve a provider credential in precedence order — Pi Agents

Calculate a context budget — Pi Agents

Read a session file and its tree — Pi Agents

See which tools Pi declares to the model — Pi Agents

Merge user and project settings — Pi Agents

Decide whether to trust a project — Pi Agents

Inspect the shell prefix and session environment — Pi Agents

Decide which Pi mechanism a requirement needs — Pi Agents

Turn a Markdown file into a /review command — Pi Agents

Advertise a skill by description, load the body on demand — Pi Agents

Give a skill files and a check script it can run — Pi Agents

Register a tool and a command from an extension — Pi Agents

Block a force push with a tool_call handler — Pi Agents

Return structured tool results and route nested calls — Pi Agents

Make what the model knows follow the active tools — Pi Agents

Persist a guard's block count as a session entry — Pi Agents

Add a slash command and a custom screen that respect the mode rule — Pi Agents

Gate MCP tools by their full mcp__ name and treat annotations as hints — Pi Agents

Ship an extension, skills, and prompts as one Pi package — Pi Agents

Take over compaction summarization — Pi Agents

Branch the session tree — Pi Agents

Steer a run without interrupting it — Pi Agents

Name the eight message roles — Pi Agents

Read a session file with the chapter's program — Pi Agents

Separate discovery from admission in the context pipeline — Pi Agents

Normalise a provider's overflow message so Pi can recover — Pi Agents

Drive one agent through three interfaces — Pi Agents

Run Pi with no person attached — Pi Agents

Read the JSON event stream record by record — Pi Agents

Embed an agent in your process with SDK sessions — Pi Agents

Drive a live Pi subprocess over JSONL — Pi Agents

Get one typed answer from one model call — Pi Agents

Run the agent loop on its own with a corpus tool — Pi Agents

Return a schema-valid value from a bounded model loop — Pi Agents

Map each agent-core hook by what it can still decide — Pi Agents

Compose a pipeline with the weakest component at each position — Pi Agents

Measure agent reliability with pass@1 and pass^k — Pi Agents

Read a self-declared annotation as a claim, not a control — Pi Agents