project-development

Name: project-development
Author: muratcankoylan

50installs1.4kforks

How to install project-development?

muratcankoylan/agent-skills-for-context-engineering/project-development

$npx -y skills add muratcankoylan/Agent-Skills-for-Context-Engineering --skill project-development

Installs into the current project.

Summary

This skill should be used for project-level decisions about LLM-powered systems: whether an LLM is the right primitive for the task at hand, the shape of a multi-stage batch or agent pipeline, token and cost estimation, choosing between single-agent and multi-agent at the project

Files · 1

View on GitHub

SKILL.md

1	# Project Development Methodology
2
3	This skill covers the principles for identifying tasks suited to LLM processing, designing effective project architectures, and iterating rapidly using agent-assisted development. The methodology applies whether building a batch processing pipeline, a multi-agent research system, or an interactive agent application.
4
5	The unit of work for this skill is the whole project or a multi-stage pipeline. Individual tool design (descriptions, schemas, error messages) belongs to `tool-design`. Per-skill activation routing belongs to the corresponding skill plus the corpus index. This skill owns the project-level questions: should you build this with an LLM at all, what shape should the pipeline take, what does it cost, how should it be iterated.
6
7	## When to Activate
8
9	Activate this skill when the unit of work is a whole project or pipeline:
10
11	- Deciding whether an LLM is the right primitive for a task at all (task-model fit before any code).
12	- Shaping a multi-stage batch or agent pipeline (acquire / prepare / process / parse / render).
13	- Estimating tokens, dollar cost, and timelines for an LLM-heavy project.
14	- Choosing between single-agent and multi-agent at the project level.
15	- Structuring agent-assisted iteration (where the agent helps build the project itself).
16	- Designing structured output at the pipeline contract level (cross-stage handoff format).
17
18	Do not activate this skill for adjacent work owned by other skills:
19
20	- Per-tool description, schema, naming, response format, error message: `tool-design`.
21	- Per-trajectory token-efficiency tactics (masking, partitioning, caching): `context-optimization`.
22	- Deciding to split work across sub-agents at the agent topology level: `multi-agent-patterns`.
23	- Designing the autonomous control loop (locked metrics, novelty gates, human approval boundaries): `harness-engineering`.
24
25	## Core Concepts
26
27	### Task-Model Fit Recognition
28
29	Evaluate task-model fit before writing any code, because building automation on a fundamentally mismatched task wastes days of effort. Run every proposed task through these two tables to decide proceed-or-stop.
30
31	Proceed when the task has these characteristics:
32
33	\| Characteristic \| Rationale \|
34	\|----------------\|-----------\|
35	\| Synthesis across sources \| LLMs combine information from multiple inputs better than rule-based alternatives \|
36	\| Subjective judgment with rubrics \| Grading, evaluation, and classification with criteria map naturally to language reasoning \|
37	\| Natural language output \| When the goal is human-readable text, LLMs deliver it natively \|
38	\| Error tolerance \| Individual failures do not break the overall system, so LLM non-determinism is acceptable \|
39	\| Batch processing \| No conversational state required between items, which keeps context clean \|
40	\| Domain knowledge in training \| The model already has relevant context, reducing prompt engineering overhead \|
41
42	Stop when the task has these characteristics:
43
44	\| Characteristic \| Rationale \|
45	\|----------------\|-----------\|
46	\| Precise computation \| Math, counting, and exact algorithms are unreliable in language models \|
47	\| Real-time requirements \| LLM latency is too high for sub-second responses \|
48	\| Perfect accuracy requirements \| Hallucination risk makes 100% accuracy impossible \|
49	\| Proprietary data dependence \| The model lacks necessary context and cannot acquire it from prompts alone \|
50	\| Sequential dependencies \| Each step depends heavily on the previous result, compounding errors \|
51	\| Deterministic output requirements \| Same input must produce identical output, which LLMs cannot guarantee \|
52
53	### The Manual Prototype Step
54
55	Always validate task-model fit with a manual test before investing in automation. Copy one representative input into the model interface, evaluate the output quality, and use the result to answer these questions:
56
57	- Does the model have the knowledge required for this task?
58	- Can the model produce output in the format needed?
59	- What level of quality should be expected at scale?
60	- Are there obvious failure modes to address?
61
62	Do this because a failed manual prototype predicts a failed automated system, while a successful one provides both a quality baseline and a prompt-design template. The test takes minutes and prevents hours of wasted development.
63
64	### Pipeline Architecture
65
66	Structure LLM projects as staged pipelines because s

People who liked this also viewed

6 picks

Type

project-development

50installs1.4kforks

View on GitHub

How to install project-development?

muratcankoylan/agent-skills-for-context-engineering/project-development

$npx -y skills add muratcankoylan/Agent-Skills-for-Context-Engineering --skill project-development

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Project Development Methodology
2
3	This skill covers the principles for identifying tasks suited to LLM processing, designing effective project architectures, and iterating rapidly using agent-assisted development. The methodology applies whether building a batch processing pipeline, a multi-agent research system, or an interactive agent application.
4
5	The unit of work for this skill is the whole project or a multi-stage pipeline. Individual tool design (descriptions, schemas, error messages) belongs to `tool-design`. Per-skill activation routing belongs to the corresponding skill plus the corpus index. This skill owns the project-level questions: should you build this with an LLM at all, what shape should the pipeline take, what does it cost, how should it be iterated.
6
7	## When to Activate
8
9	Activate this skill when the unit of work is a whole project or pipeline:
10
11	- Deciding whether an LLM is the right primitive for a task at all (task-model fit before any code).
12	- Shaping a multi-stage batch or agent pipeline (acquire / prepare / process / parse / render).
13	- Estimating tokens, dollar cost, and timelines for an LLM-heavy project.
14	- Choosing between single-agent and multi-agent at the project level.
15	- Structuring agent-assisted iteration (where the agent helps build the project itself).
16	- Designing structured output at the pipeline contract level (cross-stage handoff format).
17
18	Do not activate this skill for adjacent work owned by other skills:
19
20	- Per-tool description, schema, naming, response format, error message: `tool-design`.
21	- Per-trajectory token-efficiency tactics (masking, partitioning, caching): `context-optimization`.
22	- Deciding to split work across sub-agents at the agent topology level: `multi-agent-patterns`.
23	- Designing the autonomous control loop (locked metrics, novelty gates, human approval boundaries): `harness-engineering`.
24
25	## Core Concepts
26
27	### Task-Model Fit Recognition
28
29	Evaluate task-model fit before writing any code, because building automation on a fundamentally mismatched task wastes days of effort. Run every proposed task through these two tables to decide proceed-or-stop.
30
31	Proceed when the task has these characteristics:
32
33	\| Characteristic \| Rationale \|
34	\|----------------\|-----------\|
35	\| Synthesis across sources \| LLMs combine information from multiple inputs better than rule-based alternatives \|
36	\| Subjective judgment with rubrics \| Grading, evaluation, and classification with criteria map naturally to language reasoning \|
37	\| Natural language output \| When the goal is human-readable text, LLMs deliver it natively \|
38	\| Error tolerance \| Individual failures do not break the overall system, so LLM non-determinism is acceptable \|
39	\| Batch processing \| No conversational state required between items, which keeps context clean \|
40	\| Domain knowledge in training \| The model already has relevant context, reducing prompt engineering overhead \|
41
42	Stop when the task has these characteristics:
43
44	\| Characteristic \| Rationale \|
45	\|----------------\|-----------\|
46	\| Precise computation \| Math, counting, and exact algorithms are unreliable in language models \|
47	\| Real-time requirements \| LLM latency is too high for sub-second responses \|
48	\| Perfect accuracy requirements \| Hallucination risk makes 100% accuracy impossible \|
49	\| Proprietary data dependence \| The model lacks necessary context and cannot acquire it from prompts alone \|
50	\| Sequential dependencies \| Each step depends heavily on the previous result, compounding errors \|
51	\| Deterministic output requirements \| Same input must produce identical output, which LLMs cannot guarantee \|
52
53	### The Manual Prototype Step
54
55	Always validate task-model fit with a manual test before investing in automation. Copy one representative input into the model interface, evaluate the output quality, and use the result to answer these questions:
56
57	- Does the model have the knowledge required for this task?
58	- Can the model produce output in the format needed?
59	- What level of quality should be expected at scale?
60	- Are there obvious failure modes to address?
61
62	Do this because a failed manual prototype predicts a failed automated system, while a successful one provides both a quality baseline and a prompt-design template. The test takes minutes and prevents hours of wasted development.
63
64	### Pipeline Architecture
65
66	Structure LLM projects as staged pipelines because s

People who liked this also viewed

6 picks

Type